Citation
A Method Correlating Patient-Specific Parameters with Direct Measurements in Cadaveric Subjects to Estimate Organ Doses in Computed Tomography

Material Information

Title:
A Method Correlating Patient-Specific Parameters with Direct Measurements in Cadaveric Subjects to Estimate Organ Doses in Computed Tomography
Creator:
Sinclair, Lindsay Ann
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Biomedical Engineering
Committee Chair:
Arreola, Manuel Munoz
Committee Members:
Hintenlang, David Eric
Rill, Lynn Neitzey
Bolch, Wesley Emmett
Welt, Bruce Ari
Graduation Date:
8/10/2013

Subjects

Subjects / Keywords:
Breasts ( jstor )
Cadavers ( jstor )
Computerized axial tomography ( jstor )
Diameters ( jstor )
Dosage ( jstor )
Head ( jstor )
Human organs ( jstor )
Liver ( jstor )
Lungs ( jstor )
Skin ( jstor )
Biomedical Engineering -- Dissertations, Academic -- UF
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
Biomedical Engineering thesis, Ph.D.

Notes

Abstract:
Computed tomography has allowed for great advances in the field of diagnostic imaging. However, this progression has not ensued without careful attention to the radiation dose associated with its use. There remains a critical need to accurately assess organ doses resulting from these procedures. The ultimate goals of this research study were: first, to develop sets of empirical equations that can be utilized to calculate patient-specific organ doses for a group of commonly performed CT exams, and second, to investigate the effects of ultra-helical acquisition mode on organ doses. Both of the aforementioned goals were accomplished with the use of a standardized direct organ dose measurement methodology utilizing optically stimulated luminescent dosimeters (OSLDs) and performing the measurements on cadaveric subjects in lieu of actual patients. The OSLDs were placed on the skin and lens of the eye, and within the following organs: thyroid, brain, lungs, breasts, liver, stomach, small intestine, large intestine, uterus, and ovaries. A variety of clinically-accepted CT protocols was examined, including chest (C), abdomen (A), pelvis (P), CAP, 3-phase liver, pulmonary embolism, trauma, head, CTA head, and brain perfusion protocols.
Abstract:
Average organ doses for all body protocols examined ranged from 2 mGy to 84 mGy, with the maximum dose resulting from the three-phase liver protocol and largest detector configuration, 0.5 mm x 160. Organ dose measurement for the head protocols resulted in a dose range of 16-299 mGy, with the highest dose resulting from the lens dose of a brain perfusion protocol. Generally, organ doses were shown to increase with the size of the detector configuration. Average organ doses from 8 cadaveric subjects and five primary protocols were compiled and normalized by the exam CTDI Vol. Equation sets were derived from correlations between these dose conversion coefficients and the central effective diameter of each subject.
Abstract:
These equations present a novel and innovative method for organ dose estimation due to their origin in direct organ dose measurements. With this research, the first step has been made in acquiring the ability to calculate patient-specific organ doses in computed tomography, and in turn more appropriately estimate the risk from CT studies.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
General Note:
Adviser: Arreola, Manuel Munoz.
General Note:
INACCESSIBLE UNTIL 2015-08-31
Statement of Responsibility:
by Lindsay Ann Sinclair.

Record Information

Source Institution:
UFRGP
Holding Location:
University of Florida
Rights Management:
Copyright Lindsay Ann Sinclair. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2015
Classification:
LD1780 2013 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 A METHOD CORRELATING PATIENT SPECIFIC PARAMETERS WITH DIRECT MEASUREMENTS IN CADAVERIC SUBJECTS TO ESTIMATE ORGAN DOSES IN COMPUTED TOMOGRAPHY By LINDSAY A SINCLAIR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Lindsay A Sinclair

PAGE 3

3 To my m om a nd d ad

PAGE 4

4 ACKNOWLEDGEMENTS First and foremost, I would like to attempt to put into words the gratitude I have for Dr. Manuel Arreola. He has been my advisor, teacher, mentor, and friend. When I first began my experience at Shands he challenged me to pursue research that I was pass ionate about, and that is exactly what he helped me to do. I have a great deal of pride in this research, and none of it would have been possible without him. Whether I was in a moment of defeat or thrilled about a new idea, he was right there every minut e. For these reasons, I would like to thank him for the past four years and for the next forty, because I know that he will continue to be my mentor and friend indefinitely. Secondly, I have to thank Dr. Rill and Dr. Brateman, who have both greatly contr ibuted to my foundation as a clinical imaging physicist. Many have said this before, but I always felt that Dr. Rill had somewhat of an open door policy, whether that was her intent or not, I certainly appreciated it. They are two of the most brilliant w omen I know, and I only hope to emulate them in my future career. The first time I heard about this field, I was in the Intro to Engineering c ourse trying to figure out what to do with my interests in math and science. As soon as I heard I always wanted to be involved in medicine somehow but never had much desire to go the M.D. route; so, for me, medical physics was perfect. In an effort to learn more about the field, I spoke with Dr. Ed Dugan first. Right awa y, he introduced me to Dr. Wes Bolch, who had an enthusiasm for the field that was indeed contagious. I am so grateful for the chance he gave me as a sophomore to be involved in fascinating research. I would also like to take this opportunity to thank th e rest of my committee members, Dr. Hintenlang and Dr. Welt.

PAGE 5

5 which pushes students to think outside of the box. Dr. Welt took an immediate interest in this research and stepped in for me when I needed an e xternal committee member on a deadline. My sincerest thank you is extended towards a few people at Shands that contributed a great deal to this project, Brian Cormack, Lori Gravelle, Dr. Bidari, Dr. Peris Celda, and Mike Bickelhaup. Brian Cormack was the technologist that scanned with Tom, Anna, and myself every early morning, always with a smile and great attention to detail. This project also would not have been possible without his constant help. Lori helped form my clinical knowledge of computed tomo graphy (CT) and was always happy to answer any questions I had. Dr. Bidari performed tube placement in all of our subjects, with great attention to detail and plenty of enthusiasm for the work. Dr. Peris Celda graciously performed all head tube placemen ts, which gave us the ability to include head protocols in this research. I have very fond memories of my time spent in the office at Shands. Jen and Melissa made me feel welcome from the start, and I thank both of them for always hel ping with any issues I had. I would like to thank my fellow graduate assistants during my time at Shands, Lindsey, Bo, Monica, and Ryan. I know those relationships will continue to aid in my personal and professional development for a long time to come. I already miss my intellectual and witty conversations with Ryan Fisher. In addition, I would like to thank Tom Griglock and Anna Mench. Over the last year, Anna has given up many hours of her days and nights to help me complete this work. I am so gra teful for the support she has offered throughout this study, whether it was through one of our long conversations we had in Room 1 while waiting for CT4 to

PAGE 6

6 be free or just exchanging our latest recipes, she has been a great friend. This dissertation is ba sed on the original ideas of Dr. Tom Griglock. I will always be grateful that he allowed me to continue what he started, and did that with complete encouragement and excitement. I also consider Tom a mentor and friend, from whom I have learned a great de al I think Tom and I will forever be trying to get back to the days when we did that cool cadaver research. Finally, I have to thank my ever supportive family and boyfriend, Brandon. Brandon has played a major part in my success over the last three year s; whether it was staying up late working right by my side, turning my almost tears into laughter time and time again, or accompanying me on my drives to and from Gainesville, I am so thankful that he is in my life. I cannot wait to start this next chapte r in our lives together. Everything I am today is owed to my amazing parents. They pushed me to take every opportunity placed in front of me. My dad loves the quote by It is a perfect example of the work ethic he instilled in me. My mother is the most selfless person I know, and it is because of her I always thought that anything is possible. I thank both of them from the bottom of my heart.

PAGE 7

7 TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ .......... 11 LIST OF FIGURES ................................ ................................ ................................ ........ 14 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER INTRODUCTION ................................ ................................ ................................ .... 19 1 1.1 Motivation ................................ ................................ ................................ ......... 20 1.2 Purpose ................................ ................................ ................................ ............ 23 1.3 Specific Aims ................................ ................................ ................................ .... 23 MULTIPLE DETECTOR COMPUTED TOMOGRAPHY ................................ ......... 25 2 2.1 Ongoing Developments on the Increasing Size of Detector Arrays .................. 25 2.1.1 Clinical Applications ................................ ................................ ............... 26 2.1.2 Geometric Efficiency ................................ ................................ .............. 26 2.2 Acquisition Modes ................................ ................................ ............................. 27 2.2.1 Axial ................................ ................................ ................................ ....... 27 2.2.2 Helical ................................ ................................ ................................ .... 27 2.2.3 Volumetr ic ................................ ................................ .............................. 28 2.2.4 Ultra Helical ................................ ................................ ........................... 29 2.3 Methods of Dose Reduction ................................ ................................ .............. 29 2.3.1 Tube Current Modulation (TCM) ................................ ............................ 30 2.3.1.1 X Y TCM ................................ ................................ ................... 30 2.3.1.2 Z axis TCM ................................ ................................ ............... 31 2.3.1.3 Organ based TCM ................................ ................................ .... 31 2.3.1.4 Temporal TCM ................................ ................................ .......... 31 2.3.2 Beam Filtration ................................ ................................ ...................... 32 2.3.3 Active Collimation ................................ ................................ .................. 33 2.3.4 Iterative Reconstruction ................................ ................................ ......... 34 COMPUTED TOMOGRAPHY DOSIMETRY ................................ .......................... 37 3 3.1 Dose Indices: Multiple Scan Average Dose (MSAD)/Computed Tomography Dose Index (CTDI)/Dose Length Product (DLP) ................................ ............... 37 3.1.1 CTDI ................................ ................................ ................................ ...... 37 3.1.2 CTDI FDA ................................ ................................ ................................ 38 3.1.3 CTDI 100 ................................ ................................ ................................ .. 38 3.1.4 CTDI W ................................ ................................ ................................ .... 39

PAGE 8

8 3.1.5 CTDI Vol ................................ ................................ ................................ ... 39 3.1.6 DLP ................................ ................................ ................................ ........ 39 3.1.7 CTDI Measurement with Broad beam CT ................................ .............. 40 3.1.8 The Limitations of CTDI for Patient Dose Estimates .............................. 43 3.2 Effective Dose ................................ ................................ ................................ ... 44 3.2.1 The Most Appropriate Use of Effective Dose ................................ ......... 44 3.2.2 Methods to Estimate Effective Dose ................................ ...................... 45 3.3 Organ Doses The Gold Standard ................................ ................................ ... 46 3.4 Size Specific Dose Estimates (SSDE) ................................ .............................. 48 INSTRUMENTATION AND METHODS OF MEASUREMENT ............................... 52 4 4.1 Detection Instruments ................................ ................................ ....................... 52 4. 1.1 0.6 cc Ionization Chamber ................................ ................................ ..... 52 4.1.2 Optically Stimulated Luminescent Dosimeters (OSLDs) ........................ 53 4.1.2.1 OSL material and mechanism ................................ ................... 54 4.1.2.2 OSLD characterization ................................ .............................. 55 4.1.2.3 OSL applications in computed tomography .............................. 57 4.2 OSLD Measurement Methods ................................ ................................ .......... 57 4.2.1 Nanodot Reading and Measurement Protocol ................................ .... 57 4.2.2 Dosimeter Calibration ................................ ................................ ............ 58 4.2.2.1 Surface dosimeter calibration ................................ ................... 60 4.2.2.2 Organ dosimeter calibration ................................ ...................... 60 4.3 Measurement Mediums ................................ ................................ .................... 62 4.3.1 Physical Phantoms ................................ ................................ ................ 62 4.3.2 Computational Phantoms ................................ ................................ ...... 63 4.3.3 Cadaveric Subjects ................................ ................................ ................ 64 CHARACTERIZATION OF BEAMS IN ULTRA HELICAL ACQUISITION MODE ... 71 5 5.1 Half Value Layer ................................ ................................ ............................... 71 5.2 Beam Width ................................ ................................ ................................ ...... 72 5.3 Over Rangi ng ................................ ................................ ................................ ... 73 DIRECT ORGAN DOSE MEASUREMENT METHODOLOGY ............................... 77 6 6.1 Tube Placement ................................ ................................ ................................ 77 6.2 Organ Dose Measurement ................................ ................................ ................ 78 6.2.1 Surface Measurements ................................ ................................ .......... 79 6.2.2 ................................ ................................ ....... 80 6.2.3 Head Measurements ................................ ................................ ............. 81 6.3 CT Protocols ................................ ................................ ................................ ..... 82 6.3.1 Head Protocols ................................ ................................ ........................ 84 6.3.1.1 Head without contrast ................................ ................................ 84 6.3.1.2 CTA head ................................ ................................ ................... 84 6.3.1.3 Brain perfusion ................................ ................................ ........... 85 6.3.2 Body Protocols ................................ ................................ ........................ 86

PAGE 9

9 6.3.2.1 Chest abdomen pelvis ................................ ............................... 86 6.3.2.2 Chest ................................ ................................ ......................... 87 6.3.2.3 Abdome n ................................ ................................ ................... 87 6.3.2.4 Pelvis ................................ ................................ ......................... 88 6.3.2.5 Three phase liver ................................ ................................ ....... 88 6.3.2.6 Pulmonary embolism ................................ ................................ 88 6.3.2.7 Trauma ................................ ................................ ...................... 89 SIZE PARAMETER MEASUREMENT ................................ ................................ .... 96 7 7.1 Investigation of Size Parameters ................................ ................................ ...... 96 7.2 Methods of Measuring Size Parameters ................................ ........................... 96 RESULTS ................................ ................................ ................................ .............. 99 8 8.1 Acquisition of Subjects ................................ ................................ ...................... 99 8.2 Characterization of Beams in Ultra Helical Acquisition Mode ......................... 100 8.2.1 Half Value Layer ................................ ................................ .................... 100 8.2.2 Beam Width ................................ ................................ ........................... 100 8.2.3 Over Ranging ................................ ................................ ........................ 100 8.3 OSLD Correction Factors ................................ ................................ ................ 101 8.4 Organ Doses ................................ ................................ ................................ ... 102 8.4.1 Head Protocol Doses ................................ ................................ ............. 103 8. 4.2 Body Protocol Organ Doses ................................ ................................ .. 104 8.4.2.1 Organ doses for 4 subjects from helical and ultra helical acquisitions ................................ ................................ ............. 104 8.4.2.2 Organ doses for 8 subjects from 0.5 mm x 64 acquisitions ...... 107 8.4.3 Fetal Dose Estimates ................................ ................................ ............ 1 09 8.5 Size Specific Dose Estimates ................................ ................................ ......... 110 8.6 Size Parameter Correlations ................................ ................................ ........... 110 DISCUSSION ................................ ................................ ................................ ....... 158 9 9.1 Organ Doses from Helical vs. Ultra Helical Acquisitions ................................ 158 9.1.1 Over Ranging Effects on Organ Dose ................................ ................... 158 9.1.2 Primary Beam Organ Dose Differences between Helical and Ultra Helical Acquisitions ................................ ................................ .............. 159 9.2 Organ Doses Head Protocols ................................ ................................ ........ 161 9.3 Fetal Dose Estimates ................................ ................................ ...................... 163 9.4 Size Specific Dose Estimate Comparison ................................ ....................... 165 9.5 Size Parameter Correlations ................................ ................................ ........... 165 CONCLUSIONS ... ................................ ................................ ............................... 170 10 10.1 Summary ................................ ................................ ................................ ...... 170 10.2 Future Work ................................ ................................ ................................ .. 171 10.3 Final Words ................................ ................................ ................................ ... 172

PAGE 10

10 LIST OF REFERENCES ................................ ................................ ............................. 174 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 183

PAGE 11

11 LIST OF TABLES Table Page 4 1 Attenuation differences for a patient versus a cadaver 79 ................................ .... 70 6 1 Tube and dosimeter locations. ................................ ................................ ............ 93 8 1 Cadaver body mass index (BMI) classification. ................................ ................ 114 8 2 Half value layer (HVL) measurements (mm Al) for the AquilionONE ............... 114 8 3 Beam width and geometric efficiency. ................................ .............................. 114 8 4 Scan parameters and over ranging lengths for the head protocol at various detector configurations ................................ ................................ ..................... 115 8 5 Scan p arameters and over ranging lengths for the chest protocol at various detector configurations. ................................ ................................ .................... 116 8 6 Scan parameters and over ranging lengths for the abdomen protocol at various detector configurations. ................................ ................................ ........ 117 8 7 Calculated f factors as a function of HV Ls for each energy and filter combination used in clinical protocols. ................................ ............................. 119 8 8 O ptically stimulated luminescent dosimeter (O SLD ) su rface correction factors for each energy and filter combination utilized for adult head and body computed tomography (CT) protocols. ................................ ............................. 119 8 9 OSLD organ correction factors for each energy and filter combination utilized for adult head CT protocols. ................................ ................................ ............. 120 8 10 OSLD organ correction factors for each energy and filter combination utilized for adult body CT protocols. ................................ ................................ ............. 120 8 11 Validation of surface energy correction factors for body CT protocols. ............. 120 8 12 Organ doses for a standard head protocol at various detector configurations. 121 8 13 Organ doses for a CT angiography head protocol with a delay scan at various detector configurations. ................................ ................................ ........ 122 8 14 Brain perfusion protocol organ doses. ................................ .............................. 123 8 15 Brain perfusion protocol lens doses with and without a bismuth shield. ........... 123 8 16 Organ doses from the chest abdomen pelvis ( CAP ) protocol for various detector configurations. ................................ ................................ .................... 124

PAGE 12

12 8 17 Organ doses from the chest protocol for various detector configurations. ........ 126 8 18 Organ doses from the abdomen protocol for various detector configurations. 128 8 19 Organ doses from the pelvis protocol for various detector configurations. ....... 130 8 20 Organ doses from the three phase liver protocol for various detector configurations. ................................ ................................ ................................ .. 132 8 21 Organ doses from the pulmonary embolism protocol for various detector configurations. ................................ ................................ ................................ .. 134 8 22 Organ doses from the trauma protocol for various detector configurations. ..... 136 8 23 Percent difference in dose for various detector configurations relative to 0.5 mm x 64 average doses. ................................ ................................ .................. 138 8 24 Organ doses from a CAP protocol for 8 subjects at a detector configuration of 0.5 mm x 64. ................................ ................................ ................................ ..... 140 8 25 Organ doses from a chest protocol for 8 subjects at a detector configuration of 0.5 mm x 64. ................................ ................................ ................................ 141 8 26 Organ doses from an abdomen protocol for 8 subjects at a detector configuration of 0.5 mm x 64. ................................ ................................ ........... 142 8 27 Organ doses from a pelvis protocol for 8 subjects at a detector configuration of 0.5 mm x 64. ................................ ................................ ................................ 143 8 28 Organ doses from a three phase liver protocol for 8 subjects at a detector configuration of 0.5 mm x 64. ................................ ................................ ........... 144 8 29 Size specific dose estimate comparison with measured organ doses for cadaver 1. ................................ ................................ ................................ ......... 145 8 30 Size specific dose estimate comparison with measured organ doses for cadaver 2. ................................ ................................ ................................ ......... 146 8 31 Size specific dose estimate comparison with measured organ doses for cadaver 3. ................................ ................................ ................................ ......... 146 8 32 Size specific dose estimate comparison with measured organ doses for cadaver 4. ................................ ................................ ................................ ......... 147 8 33 Size specific dose estimate comparison with m easured organ doses for cadaver 5. ................................ ................................ ................................ ......... 147 8 34 Size specific dose estimate comparison with measured organ doses for cadaver 6. ................................ ................................ ................................ ......... 148

PAGE 13

13 8 35 Size specific dose estimate comparison with measured organ doses for cadaver 7. ................................ ................................ ................................ ......... 148 8 36 Size specific dose estimate comparison with measured organ doses for cadaver 8. ................................ ................................ ................................ ......... 149 8 37 Effective diameters for all subjects from a CAP protocol. ................................ 149 8 38 Polynomial fit equations for correlations between the effective diameter of the central slice and organ doses for a CAP protocol. ................................ ...... 151 8 39 Polynomial fit equations for correlations between the effective diameter of the central s lice and organ dose to computed tomography dose index ( CTDI Vol ) conversion coefficents for a CAP protocol. ................................ ....................... 152 8 40 Polynomial fit equations for correlations between the effective diameter of t he central slice and organ dose to CTDI Vol conversion coefficents for a chest protocol. ................................ ................................ ................................ ............ 153 8 41 Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI Vol conversion coefficents for an abdomen protocol. ................................ ................................ ............................ 153 8 42 Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI Vol conversion coefficents for a pelvis protocol. ................................ ................................ ................................ ............ 153 8 43 Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI Vol conversion coefficents for a three phase liver protocol. ................................ ................................ ......................... 154 8 44 Calculated organ doses vs. measured organ doses for a trauma protocol and an overweight subject. ................................ ................................ ...................... 154 8 45 Calculated organ doses vs. measured organ doses for a trauma protocol and an obese subject. ................................ ................................ ............................. 155 8 46 Calculated organ doses from CAP equations vs. measured organ doses for a chest protocol and a small, medium, and large subject. ................................ ... 155 8 47 Calculated organ doses from CAP equations vs. measured organ doses for an abdomen protocol and a small, medium, and large subject. ........................ 156 8 48 Calculated organ doses from CAP equations vs. measured organ doses for a pelvis protocol and a small, medium, and large subject. ................................ .. 157

PAGE 14

14 LIST OF FIGURES Figure Page 2 1 Geometric efficiency c omparison for 4 slice v. 64 slice mulitple detector computed tomography (MDCT) ................................ ................................ .......... 35 2 2 Angular and z axis tube current m odulation. ................................ ...................... 35 2 3 Illustration of active or adaptive c ollimation. ................................ ....................... 36 3 1 Graph of computed tomography dose index (CTDI) v. multiple scan average dose (MSAD) .. ................................ ................................ ................................ .... 49 3 2 Picture of CTDI PMMA phantoms as required by the FDA. ................................ 49 3 3 Comparison of dose metrics for wide b eams.. ................................ .................... 50 3 4 Dose p rofile measured by a small ionization chamber for a scan length L eq corresponding to D eq ................................ ................................ ........................... 50 3 5 Summary of methods used for size s pecific dose e stimates. Adapted from AAPM 204. 52 ................................ ................................ ................................ ....... 51 4 1 Landauer n anodo t from the InLight Microstar dosimetry s ystem (Landauer G lenwood, IL). ................................ ................................ ................................ .... 67 4 2 Simplified optically stimulated luminescent ( OSL ) reader schematic ................. 67 4 3 Surface dosimeter calibration setup. A) Ion chamber B) Nanodot ...................... 68 4 4 Organ dosimeter calibration setup for body protocols. ................................ ....... 68 4 5 CIRS anthropomorphic phantoms (Norfolk, VA). ................................ ................ 69 4 6 UF anthropomorphic phantoms. 81 ................................ ................................ ...... 69 4 7 UF library of ICRP89 reference hybrid computational p hantoms. 61 .................... 70 5 1 Computed radiography image of 0.5 x 160 mm beam w idth. ............................. 76 6 1 Tube placement system. ................................ ................................ .................... 91 6 2 Tube placement schematic. ................................ ................................ ................ 91 6 3 Dosimeter placement in the right breast, left lobe of the liver, and stomach. ...... 92 6 4 Placement of skin dosimeter strips for a chest protocol. ................................ ..... 92

PAGE 15

15 6 5 ......................... 93 6 6 Dosimeter placement for organ dose measurements in the head. ..................... 94 6 7 Scan range for standard head protocol and brain perfusion protocol. ................ 94 6 8 Scan range for CT angiography head protocol. ................................ .................. 94 6 9 Acquisition sequence for brain perfusion protocol. ................................ ............. 95 7 1 Measurement of the antero posterior (AP) and lateral dimensions within the central image for calculation of effective diameter. ................................ ............. 98 7 2 Measurement of the AP and lateral dimensions within the central slice of the following organs: A) Liver B) Colon C) Lungs. ................................ .................... 98 8 1 Thyroid doses for a standard head protocol at different detector configurations. ................................ ................................ ................................ .. 118 8 2 Smal l intestine and colon doses for a chest protocol at different detector configurations. ................................ ................................ ................................ .. 1 18 8 3 Upper lung and uterus doses for an abdomen protocol at various detector configurations. ................................ ................................ ................................ .. 119 8 4 Average organ doses for various detector configurations with a standard head protocol. ................................ ................................ ................................ ... 123 8 5 Tube current modulation plots of various detector configurati ons for the chest protocol or pulmonary embolism protocol for cadaver 5. ................................ .. 138 8 6 Tube current modulation plots of various detector configurations for the abdomen protocol or three phase liver protocol for cadaver 6. ........................ 139 8 7 Tube current modulation plots of the 0.5 mm x 64 and 0.5 mm x 160 detector configurations for the chest abdomen pelvis ( CAP ) protocol for a small (cadaver 4) and large subject (cadaver 7). ................................ ....................... 139 8 8 Fetal dose estimates for a pulmonary embolism, chest abd omen pelvis, and trauma protocol for four cadaveric subjects. ................................ ..................... 145 8 9 Organ dose vs effective diameter of the central slice for a CAP protocol. ....... 150 8 10 Organ dose vs. central effective diameter of each organ for a CAP protocol. .. 150 8 11 Organ dose to computed tomography dose index ( CTDI Vol ) conversion coefficients vs. effective diameter of the central slice of a CAP protocol. ......... 151

PAGE 16

16 8 12 Organ dose to CTDI Vol conversion coefficients vs. the central effective diameter of each organ for a CAP protocol. ................................ ..................... 152 9 1 Average organ doses in the primary scan range for the chest protocol at various detector configurations. ................................ ................................ ........ 168 9 2 Average organ doses in the primary scan range for the abdomen protocol at various detector configurations. ................................ ................................ ........ 168 9 3 Average organ doses in the primary scan range for the three phase liver protocol at various detector configurations. ................................ ...................... 169

PAGE 17

17 A bstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy A METHOD CORRELATING PATIENT SPECIFIC PARAMETERS WITH DIRECT MEASUREMENTS IN CADAVERIC SUBJECTS TO ESTIMATE ORGAN DOSES IN COMPUTED TOMOGRAPHY By Lindsay A Sinclair August 2013 Chair: Manuel Arreola Major: Biomedical Engineering C omputed tomography has allowed for great advances in the field of diagnostic imaging However, this progression has not ensued without careful attention to the radiation dose associated with its use. There remains a critical need to accurately assess organ doses resulting from these procedures. The ultimate goals of this research study were: first, to develop sets of empirical equations that can be utilized to calculate patient specific organ doses for a group of commonly performed CT exams, and second, to investigate the effects of ultra helical acquisition mode on organ doses. Both of the aforementioned goals were accomplished with the use of a standardized direct organ dose measurement methodology utilizing optically stimulated luminescent dosimeters (OSLDs) and performing the me asurements on cadaveric subjects in lie u of actual patients. The OSLDs were placed on the skin and lens of the eye, and within the following organs: thyroid, brain, lungs, breasts, liver, stomach, small intestine, large intestine, uterus, and ovaries. A v ariety of clinically accepted CT protocols was examined, including chest (C), abdomen (A), pelvis (P), CAP, 3 pha se liver, pulmonary embolism, trauma head, CTA head, and brain perfusion protocols.

PAGE 18

18 Average organ doses for all body protocols examined ranged from 2 mGy to 84 mGy, with the maximum dose resulting from the three phase liver protocol and largest detector configuration, 0.5 mm x 160. Organ dose measurement for the head protocols resulted in a dose range of 16 299 mGy, with the highest dose result ing from the lens dose of a brain perfusion protocol. Generally organ doses were shown to increase with the size of the detector configuration. Average organ doses from 8 cadaveric subjects and five primary protocols were compiled and normalized by the exam CTDI Vol Equation sets were derived from correlations between these dose conversion coefficients and the central effective diameter of each subject. These equations present a novel and innovative method for organ dose estimation due to their origin in direct organ dose measurements. With this research the first step has been made in acquiring the ability to calculate patient specific organ doses in computed tomography, and in turn more appropriately estimate the risk from CT studies.

PAGE 19

19 CHAPTER 1 INTRODUCTION Computed tomography is considered to be one of the most progressive medical technologies of the last 50 years. Since the 1979 Nobel Prize in Medicine was awarded to its inventors, CT has emerged as a critical modality in the accurate diagnosi s of many diseases. Indeed, a quote by Sir Godfrey Hounsfield in one of the first published open up a new chapter in X lutionize the field of diagnostics for years to come. 1 Computed tomography continues to advance in technology, making the role of diagnostic medical physicist paramount to the optimization of image quality, while minimizing patient dose in CT studies. Th e application of CT imaging becomes more widespread with every new development in CT technology. From cardiovascular studies to the more recent body perfusion protocols, the use of CT in varying fields of diagnostic medicine continues to expand Annual r eports on medical imaging marketing trends are published by Information Means Value (IMV) Medical Information Division, Inc. (Des Plaines, IL) Ma ny have cited a 2006 IMV CT market report, which revealed an increase of over 50 % in the n umber of CT procedu res from 1999 to 2006. 2 The most recent IMV report shows an increase of four percent when comparing CT procedures performed in 2011 to a total of 85.3 million procedures in 2012. 3 One of the foremost advancements in CT technology has been the use of mul ti channel detectors. The ability to acquire multiple slices of data per rotation drastically changed the utilit y of CT. Since 1998, facilities have been able to do CT studies acquiring 4 simultaneous slices per rotation and have progressed to the curren t 320

PAGE 20

20 slices per rotation in axial mode Recently with the introduction of ultra helical acquisition, it is possible to image up to 160 slices in helical mode marrying speed with clinical usefulness With these new innovations and the increased use of CT, more attention has recently been paid to the potential increase in radiation risks associated with CT imaging in the public media, the medical community, and among international organizations. 4,5,6 I n order to appropriately estimate the risk of radiation effects from diagnostic CT studies and evaluate the possibilities of new er CT technologies and protocols, a knowledge of organ doses is needed. 1.1 Motivation The biological effects of radiation can be either stochastic or deterministic. In the case of stochastic effects, the probability of th e effect increases with the dose received; stochastic biological effects include latent diseases, such as cancer. On the other hand, d eterministic effects requir e that the dose absorbed be above a known threshold to occur ; they i nclude erythema, cataracts and epilation among others The currently accepted risk model for s tochastic effects is a linear no threshold (LNT) model, as described by the National Acade my of Sciences reports on Biological Effects of Ionizing Radiation referred to as BEIR V and VII. 7 The LNT model is based on the idea that there is a risk of a biological effect with any amount of radiation dose absorbed by an organ or tissue. A t low ef fective doses ( less than 50 mSv), such as the ones that result from typical diagnostic imaging procedures the statistical uncertainty of the model remains a topic of serious debate This is mainly because such risk models can only be extrapolated from at omic bomb survivor data, which involves populations exposed to high whole body doses 8,9 However numerous epidemiological studies are

PAGE 21

21 underway to examine the effects from actual medical exposures. 10,11 With the increased public and media attention surro unding CT today, it is imperative that accurate dose information be generated and become available for such epidemiological studies as well as future endeavors to estimate biological risk from CT exams While it has been stated that stochastic effects have been the main concern with doses resulting from diagnostic imaging procedures the possibility of deterministic effects from CT studies still need s to be considered Deterministic effects have recently been reported in interventional radiology and ca rdiology cases, as well as some specific incidences with CT exams. 5, 12 The main organs to be considered for determi nistic effects are the skin and the lens of the eye. For interventional radiology, the Joint Commission recently issued an alert addressin g the need for preventing sentinel events, which are those procedures which result in peak skin doses over 15 Gy. 12 In the case of CT, certain organ doses, such as those to the lens of the eye must be consistently monitored, especially for patients underg oing several CT studies in a short amount of time, such as the case of a patient with cerebrovascular events, whom may undergo several brain perfusion scans over a period of a few weeks The ICRP recently revised the threshold dose for radiogenic cataract s from 2 Gy to 0.5 Gy. 13 With that being said, the benefit of these two modalities must always be taken into account when discussing the potential effects of radiation. The IR and CT procedures performed are often life saving. Recently, due to the conc erns of the possible effects from radiation exposure in all radiological studies as well as its continued expansion and increase in use, the idea of including and maintaining accurate electronic medical

PAGE 22

22 record (EMR) is bein g explored. 11 This can offer many benefits, including but not limited to: Lifetime cumulative dose records. Dose information will be included in radiology study appropriateness criteria for use by doctors ordering diagnostic procedures. It will aid the doctors in deciding which is the most appropriate study for a given patient condition in terms of patient indications This will be useful for all patients, but especially advantageous for pediatric patients, who se organs are more sensitive to radiation t han those in adults. Monitoring and improvement of radiological procedures by monitoring typical study doses. Tracking of total skin doses for more accurate assessments of potential sentinel events by providing the ability to add skin doses from various procedures to those resulting from interventional radiology cases for more accurate skin dose estimates. Refinement of current risk models, by providing the tools necessary for updating and developing new radiation risk models. Acc urate lifetime dose records will simplify the performance of retrospective studies using actual patient data from medical exposures and dose estimates that are characteristic of the technology at the time of the exam. While the benefits are clearly numero us and powerful, in order for these to happen, the current approaches to CT dosimetry must be first improved. The current parameters available to all users namely the Volumetric CT Dose Index ( CTDI Vol ) and Dose Length Product ( DLP ) are not representative of the actual patient dose following a CT study but are rather a measure of radiation output. 14 Manufacturers and clinicians, other than radiologists, make use of Effective Doses to compare potential risks among different scanners and studie s. Effective dose is defined only for a reference person, and as such should not be used for individual dose estimates. 15 Thus the primary question arises as to what is the most appropriate dose quantity for the tasks of reducing risks of stochastic effe cts and minimizing the possibility of deterministic effects. The most viable answer is the dose absorbed by individual

PAGE 23

23 organs. Organ doses offer the most specific information from in order to assess all potential biological effects They can also be ap plied to other diagnostic modalities such as nuclear medicine or interventional radiology. I n summary, accurate organ dose determination is crucial for the present and future of CT dosimetry. This dissertation serves to explore an acc urate method to measure and estimate organ doses for adult females undergoing common CT exams. 1.2 Purpose The purpose of this research was to generate empirical sets of equations that can be used to calculate patient specific organ doses resulting from a s elected group of prominent CT studies and protocols. The necessary data for the development of these sets of equations is comprised of direct dose measurements performed within a human body. In order to make these dose measurements and determinations as state of the art as possible, the secondary objective of this proj ect was to characterize the newest form of image acquisition, ultra helical scan ning mode as this mode is likely to become more common in clinical protocols in the next few years 1.3 Specific Aims In order to accomplish the overall goals of this research, the following specific aims were developed. Specific Aim 1: Characterize the ultra helical mode of acquisition on the Aquilion ONE scanner (Toshiba America Medical Systems, Tustin, CA) including measurements of beam quality and geometric efficiency Specific Aim 2 : Determine calibration factors for the optically stimulated luminescent dosimeters (Nandots, Landauer, Glenwood, IL) used in this research under CT geometry conditions Spec ific Aim 3 : Acquire several cadaveric subjects with body mass indices (BMIs) covering a clinically relevant range

PAGE 24

24 Specific Aim 4 : Perform tube placement and conduct a series of organ dose measurements for an array of CT protocols. Specific Aim 5: Analyze the results in order to develop equation set s cor relating patient specific parameters and organ doses for given CT protocols.

PAGE 25

25 CHAPTER 2 MU L TIPLE DETECTOR COMPUTED TOMOGRAPHY As mentioned in Chapter 1 the advent of multiple row detectors for CT has been one of the main contributing factors to the explosive increased use of CT in the past twenty years. The evolution of MDCT has been further enhanced by faster gantry rotation times, smaller detector wid ths, and wider anatomical coverage. These advancements have led to better temporal and spatial resolution, and in turn, expanded clinical applications of computed tomography. 2.1 Ongoing Developments on the Increasing Size of Detector Arrays Multi detector c omputed tomography (MDCT) was introduced in 1992 with a dual slice scanner. In 1998, 4 slice scanners were presented, followed by 16 slice CTs in 2001. 16 Several years later, 32 and 64 slice scanners were introduced, which remain ology departments today. In 2008, Philips (Philips Healthcare, Andover, MA) and Toshiba (Toshiba America Medical Systems, Tustin, CA) revealed a 256 slice CT, and Toshiba later introduced the Aquilion ONE, a 320 slice CT. The Aquilion ONE (AQ1) is the foc us of this research study. MDCT is characterized by multiple rows of detectors where each row allows the imaging of an individual slice of cross sectional anatomy to be imaged. Nominal beam width is often referred to as the product of N, the number of row s, or data channels, used in acquisition and T, the slice thickness of each channel. When describing an MDCT scanner, the maximum number of slices that can be reconstructed per rotation or N max is a defining parameter. These scanners can also acquire dat a by combining multiple channels in thicker slices. This can be done after data acquisition, so that a smaller number of slices are reconstructed with thicker widths.

PAGE 26

26 Clinically, there is a trade off between image noise and spatial re solution when acquir ing thin or thick slices. If the data is acquired with thin slices, greater spatial resolution is achieved, but there is also an increase in noise due to a decreased number of photons contributing to each image. Noise can be decreased with an increase in tube current, at the expense of increased patient dose. 17 2.1.1 Clinical Applications When MDCT detector arrays increased from 4 to 64 rows, applications for cardiac imaging showed the potential for a great expansion. Visualization of the c oronary arteries by MDCT imaging permitted high efficacy, non invasive procedures for cardiac patients. In addition to imaging coronary arterial branches in the sub millimeter size range, such studies allowed for determination of the composition of plaque. 18 With an ever increasing number of slices acquired per rotation, comes an increase in scan speed. Pediatric and trauma cases benefit the most from faster scan times, as in both cases, motion artifacts can become less of a concern with fast acquisitions. In some trauma cases, even a small amount of time savings can affect patient outcomes. 2.1.2 Geometric Efficiency As the number of channels used for data acquisition is increased, so is the collimated beam width, which substantially improves the geometric efficiency o f the scanner. The nominal collimated beam width is quoted at isocenter. The actual beam width at isocenter is in fact slightly larger than the nominal beam width, and it is due to arp intensity variation of the penumbra does not make it useful for image formation, thereby adding unnecessary dose to the patient. There is a relationship between beam penumbra and

PAGE 27

27 geometric efficiency. Geometric efficiency is defined as the ratio of n ominal beam width to the full width at half maximum (FWHM) of the dose profile in the z direction. As the beam width increases, the contribution of the penumbra to total x ray beam width decreases. When 4 slice CTs were released, the geometric efficiency was a concern, because only half of the beam contributed to image formation. 19 The other half was made up by the penumbra of the beam. As CT x ray beams have grown to 64 or 320 detector rows, the geometric efficiency has improved greatly. 20 This is sho wn visually by Figure 2 1. 21 2.2 Acquisition Modes 2.2.1 Axial The original mode of image acquisition in CT scanners was the axial or sequential mode. The basic scanning configuration of the original third generation scanners continues to be used today, involving a single x ray tube with an arc of detectors across from it. In these scanners, a single tube rotation was completed, followed by an increment in the patient table position along the z axis in order to perform the next axial rotation. As the x ray tube wo uld rotate once, high voltage cables would have to be unwound by performing the next rotation in opposite direction. In the early days of CT scanners, exams were limited to heads and extremities due to the slower acquisition times where motion artifacts c ould inhibit diagnosis. 17 2.2.2 Helical In helical scanning mode, the x ray tube and detector arc rotate simultaneously around the patient, while the patient table constantly translates. The path of radiation exposure forms a helix or spiral. Helical scanning mode was made possible only through the invention and implementation of slip ring technology. Slip rings electrically

PAGE 28

28 connect the stable power supply of the system with the rotating x ray tube and detector configuration, permitting continuous, uninterru pted rotations. In order to distinguish between contiguous and overlapping rotations, the term pitch was introduced for the helical scan mode. Pitch is defined as the table displacement per rotation divided by the nominal collimated beam width. A pitch o f less than one corresponds to overlapping scans. If the pitch is greater than one there are gaps between the x ray beam. A pitch of unity represents contiguous scans. 22 The major benefits of helical scanning were much shorter scan times and new options for image reconstruction. Exams could be viewed in multi ple planes (i.e. transverse, sa git t al, or coronal) without requiring additio nal exposure to the patient. 2.2.3 Volumetric A milestone in CT technology occurred with the introduction of the 256 and 320 slice scanners, capable of imaging entire organs with one rotation of the x ray tube. Volumetric imaging is analogous to an axial image acquisition but with a much broader beam and different reconstruction techniques. Cardiac and neuroimaging studies hav e benefited greatly from this increase in anatomical coverage. These scanners are capable of imaging a beating heart without generating motion artifacts and can accomplish whole brain perfusion imaging. Another advantage to volumetric scanning is the re duction in the volume of iodinated contrast required to be administered for a given exam due to the faster acquisition times for extended anatomical coverages. 23 Furthermore, patients with a cardiovascular condition that may have a contraindication to bet a blockers are now able to have CT cardiac imaging studies completed without restrictions.

PAGE 29

29 2.2.4 Ultra Helical Within a few years of the inception of MDCT, 64 slice scanners were thought to be the pinnacle of helical scanning in diagnostic imaging; the complexities of beam controlling and reconstruction algorithms appeared to be insurmountable. Recently, however, a new form of image acquisition has been made possible with wide volume scanners. A new generation of image acquisition termed ultra helical scanning allows the speed of helical acquisition to be combined with wide coverage. Detector configurations for ultra helical mode consist of 0.5 mm x 80, 0.5 mm x 100, and 0.5 mm x 160 with corresponding nominal beam widths of 40, 50, and 80 mm. Increase s in scan speed can lead to a plethora of clinical applications in a similar fashion as the advancements implemented with the advent in scanner technology from 16 to 64 slice scanners. While the increase in clinical functions is of great benefit, the dose to the patient needs to be considered, and methods of dose reduction thoroughly investigated, as it must be the case with all new x ray imaging technology. Options for dose reduction in ultra helical scanning mode include tube current modulation and a be tter geometric efficiency of wide beams. However, the possibility of higher organ doses could result from over ranging effects associated with conventional helical scans. 24 In order to clinically optimize protocols with ultra helical acquisitions, organ doses for this acquisition mode were measured, as well as, with the standard helical acquisition mode. The balance of dose reduction and clinical optimization was examined for ultra helical acquisitions. 2.3 Methods of Dose Reduction While the use of CT has increased significantly since its inception, the corresponding dose per exam has shown a consistent decrease. 25 Major contributions

PAGE 30

30 to this trend include the use of tube current modulation, improvements in geometric efficiency, and the development of protocol optimization. 2.3.1 Tube Current Modulation (TCM) Tube current modulation (TCM) was integrated into CT in 1994. 26 It was a major step in dose reduction for the modality and that still holds true today. Tube current modulation is based on the premis e that x ray attenuation varies greatly depending on the patient thickness and what part of the body is being exposed. The implementation of TCM is done differently by different manufacturers, but tube current values are usually adjusted for a pre determi ned level of noise. Some scanners base the modulation off of the scout images making use of a preprogrammed algorithm; while others modulate real time by means of a continuous feedback mechanism. In all cases, careful attention must be paid to correctly centering the patient within the gantry. If this is not done, and the patient is not centered on the vertical axis, patient size can be misinterpreted by the scanner due to apparent positive or negative magnification. 27 There are several types of tube cu rrent modulation available on current scanners. The most common method is to combine x y TCM and z axis TCM to increase effectiveness, as shown in Figure 2 2. 2.3.1.1 X Y TCM X y tube current modulation is also referred to as angular modulation as the TCM algorit hm adjusts the mA based on the projection angle of the ray entering the patient. The mA is changed throughout a given rotation in order to reduce the dose to areas of less attenuation, while maintaining adequate and consistent signal to noise (SNR) levels (and therefore photon fluence) for the entire rotation. For example, when the tube is positioned over the top of the patient in the AP direction, a lower mA is required

PAGE 31

31 than when the tube is going laterally through the patient. If the tube current is no t increased for the lateral projection angles, low SNRs and even photon starvation can occur through highly attenuating areas, such as the shoulders. 2.3.1.2 Z axis TCM Another form of tube current modulation is applied along the z axis, with the tube current b eing modulated along the superior inferior length of the patient. The modulation alters the mA for different body regions, such as the abdomen and bony pelvis, and it is determined from the lateral scout image. In this case, all manufacturers utilize thi s predetermined form of TCM along the z axis. 2.3.1.3 Organ based TCM The purpose of organ based tube current modulation is to reduce dose to specific radiosensitive organs, such as the breast, thyroid, and lens of the eye. From the scout images and the orienta tion of the patient in the scanner, the organ based TCM algorithm determines the particular z locations and approximate angular projections which will encompass these organs. Thus, the scanner reduces the tube current for a specific angle distribution wit hin each tube rotation. For the organs listed above, the anterior portion of a rotation is completed with a 75 90 percent reduction in tube current, while the lateral and posterior angles of projection are associated with an adequate increase in tube curr ent. This kind of modulation therefore requires the use of a reconstruction algorithm which preferentially weights the projections done with a higher mA in order to maintain image quality. 28 2.3.1.4 Temporal TCM Tube current can also be modulated as a function o f time as a scan progresses in a protocol specific manner, such as the case of TCM ECG gated cardiac studies. In

PAGE 32

32 this type of study using temporal TCM, the cardiac cycle (through the ECG signal) is used as a guide to decrease tube current during systole, when the contraction of atria and ventricles of the heart prevents generating motion free images of diagnostic quality. 29 Likewise, prospective ECG triggering protocols with volume acquisitions have shown the greatest dose reduction for coronary CTA stud ies. In this case, t ECG is still used but instead of only modulating the tube current, the x ray tube essentially shuts off during systole, saving patient exposure. 30 A similar application is the one utilized for helical chest CTA protocols. Further do se reduction for this exam has been studied at institutions with wide volume scanners. 2.3.2 Beam Filtration The ultimate goal of filtration is to remove soft or less penetrating x rays from the beam in an effort to decrease surface dose. Computed tomography ma kes use of specific beam filters. In addition to a standard Cu flat filter which simply removes the lower portion of the x ray spectra which would only contribute to patient skin dose (an issue of considerable importance for certain studies, such as brain perfusion), a bow tie filter (named so because of its shape) is also utilized. While the x ray tube rotates around the patient in CT, the center of the beam consistently encounters the thicker part of the anatomy. In order to compensate for this differe nce in overall attenuation and beam hardening, the shape of the bow tie filter starts out thick on the outside and then gradually decreases in thickness towards the center. As a result, the dose to the periphery is decreased while maintaining SNR, and the refore image quality, in the center.

PAGE 33

33 There are three bowtie filters of different sizes available on the 320 slice scanner. The scanner selects a bowtie filter corresponding with the scan field of view or FOV chosen for a specific patient and study. 2.3.3 Ac tive Collimation For accurate and reliable reconstruction of images from helical acquisitions, additional projection data is needed outside of the planned anatomical scan range in order to properly reconstruct the first and last images of a scan set. Thus it is necessary for the beam to perform additional rotations of the CT beam both pre and post the selected anatomical scan range. This is referred to as over ranging. Over ranging results in irradiation of organs outside of the anatomical area being imaged. This concept can also be described in terms of the over ranging length or the extra length irradiated outside of the planned scan length. Over ranging effects on patient dose have been shown to worsen with increasing pitch and beam width. 31 A ctive collimation is the current solution to the problem of exposure due to over ranging. Dynamic collimators are used at the beginning and end of a helical scan to limit the full width of the beam from irradiating the patient. An illustration of adaptiv e collimation is shown in Figure 2 3. In a study done at the Mayo Clinic (Rochester, MN), dose reduction from active collimation on a 64 MDCT was shown to be anywhere between 3 46 %. The greatest reductions in dose were observed in exams with short scan lengths and higher values of pitch. 32 The effects of over ranging and the use of active collimation for ultra helical acquisitions have not yet been investigated. The potential clinical advantages of ultra helical scanning are clear, as faster and wider coverage may allow for faster studies. Since the dose effects appear to be exacerbated with wider collimations, further attention must be paid to this issue. This project served

PAGE 34

34 to examine over ranging effects with ultra helical detector configurations, and is discussed further in Chapter 5. 2.3.4 Iterative Reconstruction Though not a new concept in CT, the most recent method of dose reduction is the use of more efficient iterative reconstruction algorithms slowly phasing out traditional filtered backprojection. Iterative reconstruction had been used in the early days of computed tomography, but the computing power at the time did not allow for it to prosper. 33 Most reconstruction algorithms currently utilize some comb ination of iterative and fil tered back projection (FBP) methods. The evaluation of image quality still needs to be determined as iterative methods handle noise differently than traditional FBP. The use of these new reconstruction methods will be examined in future studies that are beyond the scope of this project.

PAGE 35

35 Figure 2 1. Geometric efficiency c omparison for 4 slice v. 64 slice multiple detector computed tomography ( MDCT ) Adapted from Rogalla P, Kloeters C, Hein PA. CT Technology Overview: 64 slice and Beyond. Radiol Clin N Am 2009; 47:1 11. Figure 2 2. Angular and z axis tube current m odulation. Adapted from McCollough CH, Bruesewitz MR, Kofler JM. CT Dose Reduction and Dose Management Tools: Overview of Available Options. Radiology. March 2006; 26:503 512.34

PAGE 36

36 Fi gure 2 3. Illustratio n of active or adaptive c ollimation. Adapted from Deak PD, Langner O, Lell M, Kalendar WA. Effects of Adaptive Section Collimation on Patient Radiation Dose in Multisection Spiral CT. Radiology. July 2009; 252: 140 147.35

PAGE 37

37 CHAPTER 3 C OMPUTED TOMOGRAPHY DOSIMETRY The computed tomography dose index is a dose quantity that is widely known and is included in federal regulations. This chapter serves to exp lore the history of CT dosimetry including the evolution of the CTDI, available pati ent dose descriptors, and how those doses can apply to radiation risk. 3.1 Dose Indices: M ultiple S can A verage D ose (MSAD) /C omputed T omography D ose I ndex (CTDI) /D ose L ength P roduct (DLP) 3.1.1 CTDI The original standardized dose metric for CT was the computed tomography dose index, or CTDI, developed in 1981. 36 The equation below represents the first version of the dose descriptor, where T is the slice width, and the function D(z) represents the dose profile as a function of position on the z axis. ( 3 1) CTDI was developed as an estimate of the multiple scan average dose (MSAD). The MSAD is measured by placing a dosimeter in the center of a phantom and scanning the phantom with contiguous axial slices. The MSAD is then cal culated by summing the contributions of each scan to the central slice. The CTDI provided a quick and accurate estimation of the MSAD, by measuring the dose profile along some length with one axial scan. The division of the dose profile by the slice thick ness, T and later the nominal beam width NT, ensures that the CTDI is an appropriate approximation of the MSAD in that the scatter tails outside of T are included in the dose. This holds as long as the spacing between slices is equal to the slice thicknes s. 36 The SI units for CTDI are milligrays or mGy.

PAGE 38

38 3.1.2 CTDI FDA The first alteration to CTDI was conducted by the U.S. Food and Drug Administration. The integration limits were changed on the original CTDI from infinity to 7T and the total beam width was cha nged to include N, the number of data channels, given that the first CT scanner was actually a dual slice scanner. 37,38, 39 ( 3 2 ) At this time, the FDA also standardized the measurement media for CTDI. They approved 2 cylindrical phantoms, which were 14 to 15 cm in length and either 16 or 32 cm in diameter. The 16 cm phantom was meant to represent an adult head, while the 32 cm phantom was representative of an adult torso. Both phantoms were made out of acrylic or polymethylmethacrylate (PMMA), a material chosen for its similar attenuation to patient soft tissue. The CT acrylic phantoms used today fo llow the same specifications. 3.1.3 CTDI 100 The next version of CTDI, CTDI 100 is still used today and was an effort to further standardize CT dosimetry. The integration limits of 14T proved to be too narrow as the slice width became smaller and the number of data channels increased, not allowing for all of the dose profile to be c aptured. Changing the integrated length to 100 mm allowed for CTDI measurements across manufacturers to be structured. ( 3 3 ) It is measured with a 100 mm pencil ion chamber, consisting of a 3 cm 3 active volume. 40 41

PAGE 39

39 3.1.4 CTDI W CTDI 100 can be measured at either the center or periphery of the phantom. The value of CTDI 100 will vary based on the measurement location due to the differing amounts of attenuation experienced. The weighted CTDI or CTDI W combines the center and peripheral measurements. Incorporating both positions allows for a better estimation of average dose in the x y plane. 42 ( 3 4 ) 3.1.5 CTDI Vol With the advent of helical or spiral scanning an additional factor of pitch was added to account for non contiguous slices, creating the term, CTDI VOL Pitch is defined as the table increment per rotation (I) divided by the nominal beam collimation. 38 ( 3 5 ) It is the CTDI Vol that is the standard CT dose metric currently required for all protocols on existing CT units. 42 3.1.6 D LP The CTDI Vol as well as another dose descriptor, the dose length product (DLP) accompany protocol selections on scanners. The equation for DLP is shown below. ( 3 6 ) In terms of a dose index, the DLP is an impr ovement on CTDI, due to the incorporation of the scan length (L) of an exam. The scan length can vary from patient to patient for the same protocol. This variation is a result of anatomical differences among patients. Hence, the DLP is in part, a patien t specific indicator.

PAGE 40

40 3.1.7 CTDI Measurement with Broad beam CT Since 1994, the number of detector rows in MDCT has steadily increased. With the corresponding increase in beam width, the CTDI concept has been reexamined by multiple studies. Dixon found that the 100 mm pencil ion chamber underestimated the CTDI in a body phantom for a 20 mm beam width by twenty percent. 43 It has been established that the 100 mm pencil ion chamber is no less efficient in measuring the dose profile for up to a 40 mm nominal bea m width. 44 However, at beam widths greater than 40 mm, a larger amount of the scatter contribution will be excluded in a CTDI measurement. When the 256 and 320 slice MDCTs came on the market, it became clear the current definition and use of CTDI would no longer be valid for these models. The nominal beam width is wider than the actual phantoms and pencil ion chambers used for CTDI measurem ents. For the AQ1, the nominal beam width at isocenter is 160 mm and is 128 mm for the 256 slice scanner. There has been several methods proposed to resolve this issue including but not limited to longer phantoms, smaller ion chambers and new dose metric s. One method is to use several phantoms adjacent to one another, essentially lengthening the measurement medium to reach the equilibrium dose, in conjunction with a small volume ion chamber to measure the dose at the center of the phantom. 43 In a study of dose profiles by Mori et al., it was established that in order to encompass more than 90 percent of the dose profile for varying beam widths up to 128 mm (for a 256 slice scanner), the phantom needed to be larger than 300 mm. 45 Mori then published con version factors based on measurements with a 300 mm ion chamber and the 350 mm phantom to be applied to the standard CTDI measurement techniques

PAGE 41

41 with a 100 mm pencil ion chamber. The conversion factors are categorized by beam width and effective energy so that they can be applied to CTDI W for other scanners. The author encourages the use of the conversion factors even for beams smaller than 100 mm given the inefficiency of CTDI 100 as it is measured now. 46 When the 256 slice prototype scanner was updated t o a 320 slice scanner for Toshiba, another study was done with the same methodology discussed above with the addition of new dose metrics. The study proved with experimental measurements that CTDI 100 greatly underestimated the dose profile of the 160 mm b eam. The authors implemented to describe the average dose measured by the 100 mm ion chamber from a wide beam acquisition. Since most clinical uses of the 0.5 x 320 detector configuration will be for a single scan, this dose quantity is more appropria te than the standard CTDI which measures the average dose from multiple scans. Finally, while the preferred quantity is the average dose, correction factors are given to enable calculation of CTDI 300,W from 100 Figure 3 3 shows schematics of this mea surement methodology. 47 In AAPM Report No. 111, new dose parameters are explored. It is proposed to measure the equilibrium dose within a phantom and conduct free in air measurements for every set of scan techniques available on a scanner. These measurem ents are suggested to be completed in a phantom of least 450 mm in length and with a thimble ion chamber so as to measure the dose at z=0 on the longitudinal axis. Figure 3 4 illustrates the measurement of D L (z=0) from which the equilibrium dose can be e acquisition used clinically. For example, solutions are suggested for stationary cone

PAGE 42

42 beam CT exams such as perfusion studies, an area that is not covered with curren t CTDI methodology. This method appears promising, however several details about specific phantoms and what media the dose values would be reported to still need to be worked out. 14 Lastly, the IEC has adopted a method to scale the CTDI by ratios of free i n air measurements. For nominal beam widths less than or equal to 40 mm, CTDI 100 is calculated in the same way as it was before, shown in equation 3. For nominal beam widths greater than 40 mm, the CTDI 100,NxT is calculated by multiplying a CTDI 100,ref by a ratio of free in air CTDI measurements for the wide beam width being considered, as well as the reference beam width. This is shown in Equation 7. The reference beam width is 20 mm or the next smallest collimation. 48 ( 3 7 ) Free in air measurements for beam widths less than or equal to 60 mm, an integration length of at least 100 mm is required. For beam widths greater than 60 mm, an integration length of at least NxT + 40 mm is required. This can be accomplished by two contiguous measurements with the pencil ion chamber or other adequate ion chambers, larger or smaller. 49 The purpose of adopting this methodology is t o make CTDI efficiency constant for all beam widths. An admitted weakness of this approach is that the CTDI 100 still substantially underestimates the equilibrium CTDI value, that which would be obtained in a phantom of infinite length. Furthermore, the C TDI concept overestimates the dose from single wide beam acquisitions. Several organizations are currently reviewing these inadequacies, including AAPM Task Group No. 200 and the International

PAGE 43

43 Commission on Radiation Units and Measurements (ICRU). These groups are making an effort to improve upon CT dosimetry, while still keeping the regularity that comes with the concept of CTDI. 3.1.8 The Limitations of CTDI for Patient Dose Estimates CTDI Vol as well as its predecessors were intended to be a standard mea sure of radiation output from a CT scanner. This standardization proved to be a significant contribution to the field of CT dosimetry. It allowed the radiation output of CT to be a reproducible, reliable, and accessible measure. As stated previously, fe deral regulations require CTDI and DLP to be built in to all protocols on a scanner. A great advantage of these metrics is that they allow for comparisons between protocols on one scanner or comparison of radiation output on multiple scanners. Due to it s accessibility, many have attempted to relate CTDI to patient dose. A direct comparison to patient dose is frowned upon in the medical physics community. While the dose length product accounts for a specific scan range, it is still calculated from CTDI Vol which can significantly underestimate or overestimate patient dose. For instance, if a fairly slim adult or pediatric patient is being imaged, and the scanner calculates CTDI Vol for the 32 cm phantom, this can be a great underestimation of the patient for pediatric patients in that they are more radiosensitive than adults. Conversely, the same phantom will be used to estimate the CT dose index for an obese patient. In this case, the dose is overestimated, as more radiation is attenuated by the larger patient effectively shielding many of the internal organs. 50 Due to the numerous issues with relating CTDI directly to patient dose, most research efforts are dedicated to a more ind irect relationship. These efforts take

PAGE 44

44 advantage of the feasibility of obtaining CTDI or DLP, but also make efforts to account for patient size and more advanced forms of technology not accounted for in CTDI calculation. 51,52 3.2 Effective Dose The concept of effective dose was introduced by the ICRP in an effort to establish radiation protection quantities from which dose limits could be set for radiation workers and members of the general public. 53, 54, 15 The primary objective of the effective dose is to e quate a non uniform partial body irradiation to a whole body exposure. In normalizing to whole body exposures, comparisons of risk can be made. Equation 3 7 represents the most recent definition of the effective dose from ICRP 103. 15 ( 3 8 ) In this equation w T is the tissue or organ weighting factor, which takes the radiosensitivity of an organ into account. In the second summation, w R represents the radiation weighting factor. The purpose of this weighting factor is to accoun t for the relative biological effectiveness of a particular radiation type in producing stochastic effects at low doses. 3.2.1 The Most Appropriate Use of Effective Dose Effective dose is defined for use in radiation protection to establish regulations and reference levels in a prospective manner. A retrospective analysis of effective dose could be required if for instance, a radiation worker surpasses dose limits. In th is case, the effective dose calculation is instructed to be used as a broad estimation of risk, but specific organ doses are to be used for a more in depth analysis of risk. ICRP 103

PAGE 45

45 benefit ass essments, the 15 Effective dose is not an appropriate measure of medical exposures or risk for any individual patient for several reasons. The tissue weighting factors u sed in the calculation of effective dose are age and sex averaged for a reference person. This averaging is an indication that the values are to be used for a reference person and not for an individual. The ICRP advises against using effective dose for individual dose calculations as well as for epidemiological studies. The effective dose can be helpful in comparing procedures to some reference quantity, such as background radiation. It can also be used to compare doses across various diagnostic modali ties. For these reasons, there have been many attempts to quickly estimate effective dose, however organ doses remain the most effective quantity by which medical exposures should be characterized. 3.2.2 Methods to Estimate Effective Dose There are two main way s in which effective dose can be calculated. The first is that recommended by the ICRP, in which organ doses are computed from Monte Carlo simulations for a reference person and ICRP weighting factors are applied. The second involves a widely known effor 55 These conversion factors are based on DLP data from a wide variety of scanners in the UK, as well as Monte Carlo organ dose data from 5 different phantoms, incl uding 4 pediatric phantoms and 1 adult phantom. 25, 56 While the intent of these conversion factors is to provide a broad quick estimate of effective dose for region specific CT exams, many studies inaccurately apply them to individual patient dose and ris k estimates. 57 Reasons why effective dose should not be

PAGE 46

46 used for patient dose estimation still apply for the conversion factors. In addition, the weighting factors used to calculate the conversion factors are not based on the most recent ICRP report; ins tead, they are based on the tissue weighting factors from ICRP 60. 15 3.3 Organ Doses The Gold Standard Organ dose measurements provide a duality for examining the effects of radiation. They can be used to estimate deterministic or stochastic effects of ra diation. They can also be used to examine individual patient doses, as well as for future epidemiological studies. There exists a wide variety of uses for organ doses. While the focus of radiation risk in diagnostic imaging is stochastic in nature, rece ntly deterministic effects are being considered. On April of 2011, the ICRP released a statement stating that the threshold for absorbed dose to the lens of the eye is now 0.5 Gy, a substantial reduction from the previous value of 2 Gy. 13 This informatio n presses the need for absorbed dose estimates to the lens from CT scans of the head, particularly brain perfusion scans. In interventional radiology, lengthy procedures can lead to sentinel events. For these high dose procedures, peak skin dose is calculated and reported. In order to fully encompass the dose accumulated by the patient, CT scans need to be accounted for as well. If a maximum or average skin dose was recorded with an exam and made readily available, it could be used in these situations. Current methods for measuring organ doses from computed tomography can be divided into 2 categories: computer simulations utilizing radiation transport codes and computational phantoms, and experimental measurements using physical phantoms and a selected dosimetry system. A few of the most prevalent computer programs for

PAGE 47

47 patient dose calculation include the IMPACT CT patien t dose calculator 58 and CT Expo. 59 While this software is the most robust and user friendly out there, the phantoms used in these simulations are first generation stylized phantoms. Current hybrid phantoms, that utilize NURBS (non uniform rational B spli ne) surfaces, are more anatomically realistic and flexible to specific patient measurements. 60, 61 In one study conducted by Turner et al., CTDI vol was used as a normalization factor to achieve scanner independent organ doses. For this study 4 MCNP models of a 64 slice CT scanner were used, corresponding to 4 different manufacturers. Simulations were completed on the GSF adult female voxel phantom. The purpose was to examine the feasibility of developing CTDI Vol to organ dose conversion coefficients th at are scanner independent. The study states that the work needs to be expanded to include a larger number of phantoms and specific CT protocols. 62 Huda et al. promotes the use of in patient to isocenter kerma ratios to deduce organ doses from CT exams. to measurements done in a Rando phantom utilizing thermoluminescent dosimeters on 2 different scanners. It is also mentioned that the ratios can only be applied to a patient of similar size to th e phantom and corrections must be made for exams other than a whole body CT. 63 In summary, estimations of organ doses can be made in a variety of ways. Compromises must be made with all the methods utilized in order to replicate current CT technology acc urately and to create a model of the patient with realistic anatomy. This project measured actual organ doses for adult females with a range of BMIs for a range of CT pr otocols.

PAGE 48

48 3.4 Size Specific Dose Estimates (SSDE) The idea of size specific dose estimates was recently int roduced by AAPM Task Group 204. (AAPM 204) Data from various sources including Monte Carlo simulations, physical measurements with anthropomorphic phantoms, and physical measurements with cylindrical acrylic phantoms were pooled together to create a library of conversion estimate. Figure 3 5 shows the various phantoms and simulation methods used for these dose estimates. This report includes a set of con version factors based on 1 or 2 manual measurements on a CT image. The conversion factors are applied to an exam CTDI Vol with which a patient specific dose estimate can be calculated. 52 This method allows for quick estimation of a patient specific dose, however, it is based on CTDI Vol and represents the peak dose along the z axis. Hence, it is not indicative of specific organ doses, does not address head exams, and does not account for the shortcomings of the CTDI in capturing all of the scatter tails of a beam. 64 Although, the SSDE protocol does offer an easily accessible solution while organ dose libraries and programs are being developed. Various studies have recently shown the applicability of this dose quantity in the clinic. 65 This document inc ludes a comparison between direct organ dose measurements and size specific dose estimates.

PAGE 49

49 Figure 3 1. Graph of computed tomography dose index (C TDI ) v. multiple scan average dose ( MSAD ) Adapted from Shope, T., R. Gagne, and G. Johnson. (1981). ray computed Med Phys 8:488 495. Figure 3 2. Picture of CTDI PMMA phantoms as required by the FDA.

PAGE 50

50 Figure 3 3. Comparison of dose metrics for wide b eams Adapted fr om J. Geleijns, A.M. Salvado, P.W. de Bruin, R. Mather, Y. Muramatsu, and M.F. Nitt Gray, "Computed tomography dose assessment for a 160 mm wide, 320 detector row, cone beam CT scanner," Phys. Med. Biol. 54 3141 3159 (2009). Figure 3 4. Dose p rofile m easured by a small ionization chamber for a scan length L eq corresponding to D eq Adapted from AMERICAN ASSOCIATION OF PHYSICISTS IN MEDICINE, Comprehensive Methodology for the Evaluation of Radiation Dose in X Ray Computed Tomography, AAPM Rep. 111 New Yo rk (2010).

PAGE 51

51 Figure 3 5. Summary of methods used for size specific dose e stimates. Adapted from AAPM 204. 52

PAGE 52

52 CHAPTER 4 INSTRUMENTATION AND METHODS OF MEASUREMENT A variety of instruments were used in this project to measure air kerma and absorbed dose, in conjunction with a variety of phantoms and x ray attenuating media. Point organ doses were measured for this research using optically stimulate d luminescent Nano dot dosimeters (Landauer Inc., Glenwood, IL) Air kerma measurements were performed using Radcal 10X6 6 and 10X6 0.6 ion chambers (Radcal, Monrovia, CA). Organ doses were measured directly on cadaveric specimen s, while a variety of phantoms, such as the CTDI PMMA cylindrical phantoms, were used with Nanodots and ion chambers The accurate use of these measurement tools and parameters are described in this chapter. 4.1 Detection Instruments 4.1.1 0.6 cc Ionization Chamber When considering requirements for an ideal radiation detector the following aspects must be examined: reproducibility, stability, linearity, angular dependence, energy depe ndence, size, cost, and convenience of use. Ion chambers have been used for decades because they meet most of these ideal con ditions. 66 Specifically, t he 0.6 cc small volume ion chamber will be utilized in this project as the gold standard in dose measurement. This approach follows the recommendations of AAPM Report No. 111, which indicates that a Farmer chamber is the newly p roposed method of physica l dose measurement for CT dosimetry. A Farmer chamber is defined as having an active volume of 0.6 cc and an active length of 20 35 mm. The small size of the ion chamber al lows for accurate measurement at a single position (z=0) o n the z axis. 14

PAGE 53

53 The 0.6cc ion chamber used for this study was calibrated according to the National Institute of Standards and Technology (NIST) standard for a M120 beam quality The M120 beam quality is defined as 12 0 kVp, 6.79 mm Al half value layer and was therefore selected to be the beam of calibration as it closely matches the typical 120 kVp beam of the AQ1 320 slice scanner 67 The clinical relevance of such a calibration beam comes from the fact that such is the beam of choice for most of the clin ical protocols used in this study and overall, the most widely used clinically. The appropriate correction factor provided by the calibration lab was applied to all measurements made with the ion chamber. 4.1.2 O ptically S timulated L uminescent Dosimeters (OSLD s) The utilization of optically stimulated luminescence for dosimetry purposes began background radiation. 66 One major advantage of using OSL dosimetry is the high radi ation sensitivity associated with Al 2 O 3 :C. Doses on the order of a few micrograys can be accurately and reproducibly detected with OSLs This makes the material an excellent choice to measure backgrou nd radiation, well below the reliable levels for thermoluminescent dosimeters ( TLDs) 68 Due to their high sensitivity OSL dosimeters (OSLDs) can be manufacturered into very small form (thus the name N anodots given by Landauer (Glenwood, IL) to measure doses in diagnostic radiology. They are often comp ared to TLDs because of their common ality in the materials used for them processing techniques, and small size. The most common OSL dosimeter material, Al 2 O 3 was originally used as a TLD. While simple, rapid readout and erasure is completed with exposure to UV light for a certain

PAGE 54

54 period of time. Each readout takes a few hundred milliseconds. This short processing time yields feasibility of large scale dosimetry operations, li ke the one explored in this p r oject 68 4.1.2.1 OSL material and m echanism Some materials that have been used for their OSL properties include Al 2 O 3 :C, BeO, quartz and MgO:Tb. 69 As mentioned earlier, the most widely known and commercially available OSL materia l is Al 2 O 3 :C ; which Landauer (Glenwood, IL) uses for their Nanodot dosimeters. Thus, the discussion of OSLDs from this point forward corresponds to aluminum oxide OSLs In fact, It is now the main material from which pe rsonnel dosimeters are constructed (Luxel, Landauer, Glenwood, IL) Nanodots have recently found a role in assessing doses in diagnostic imaging via the Microstar InLight Dosimetry System, made commerc ially available by Landauer (Glenwood, IL) The Microstar system consists of Nanodots , holders for the N anodots a laptop and a dosimeter reader. The N anodot dosimeter is a 5 mm diameter, 0.2 mm thick disk of Al 2 O 3 :C inside a light tight plastic case with dimensions of 10x10x2 mm 3 The N anodot is shown in Figure 4 1. The particular dosimeters used f and a manufacturer stated accuracy of + 2%. 70 When the OSL material is irradiated, atoms in the crystal are excited and electron/hole pairs are created. The carbon d oping of the crystal lattice results in electrons or holes to be trapped in excited energy states following excitation. Following irradiation, the Nanodot is placed in a holder which is then inserted into the reader. Once inside the reader, the actual O SL is removed from its casing and stimulated with an LED (light emitting diode) emitting a wavelength of 540 nm. This light raises the

PAGE 55

55 trapped electrons to the luminescence level and the OSL emits light with an emission band centered at 415 nm which is co llected by a photomultiplier tube (PMT) The luminescence, and thus the output from the PMT is proportional to dose received by the Nanodot. There are filters in between the stimulating source and dosimeter, as well as, in between the transducer (PMT) a nd dosimeter. This is to en sure that only the luminescence of the wavelengths emitted by the OSL are released and collected. 69 The configuration of the reader is shown in Figure 4 2. 4.1.2.2 OSLD c haracterization There have been several studies documenting the ch aracterization of OSL dosimeters in the diagnostic and therapeutic energy range. 71, 72, 73, 74 The properties examined include linearity, reproducibility, angular dependence and the known energy dependence associated with Al 2 O 3 :C. This section will high light the findings of these studies for the diagnostic energy range. The initial characterization of N anodot dosimeters was done in 2007 by Jursinic et al. for an energy range primarily within the range of radiation therapy The study reports a linear response of the dosimeter or doses up to 3 Gy and reports a supra linear reponse above 3 Gy. 71 In the diagnostic energy range, Lavoie et al. observed a linear response from 0 mGy 1000 mGy. 72 Another study found linearity over doses for three different di agnostic modalities: mammography, general radiography, and computed tomography. 73 Regarding reproducibility, Lavoie et al. found a 2.3% coeff icient of variation for Nanodot dosimeters which is very close to the manufacturer sta ted accuracy of 2% 72 The OSLD angular dependence has been shown to become evident and increase with lower energies. In a study done by the University of Texas Health

PAGE 56

56 Science Center, the maximum drop in response was at 90 deg rees with a mammographic beam of 25 kVp. For genera l radiography, the drop in response observed at 90 degrees at 80 and 120 kVp was 40 and 20 percent, respectively For similar beams in CT only a 10% drop in response at 90 degrees is reported The drop in response at the 90 degree angle is due to the geo metrical shape of the dosimeter as well as to thicker part of the casing encountered by photons from that angle. The 1mm casing is equivalent to water in mass density and could attenuate low energy photons to a certain extent. 73 In the University of Flori da study, the maximum drop in OSL re s ponse for in air measurements occurred at 90 degrees as well with a 22 % drop. For in phantom measurements, a maximum drop of 4% was observed but proven to not be statistically significant. 72 Angular dependence was c onsidered in this study, however since no measurements were taken solely in air no corrections for angular dependence were made. While the OSLDs show angular independence i n a phantom and a linear and reproducible response, the main caveat to OSLD use at diagnostic energy levels is in fact its energy dependence. The dosimeter material tends to over respond in tissue due in part to an effective Z of 11.28 for Al 2 O 3 :C. 75 The energy dependence signifies that the response of the dosimeters will change with tube potential and other factors that alter the spectrum and therefore the effective energy of the beam. The application of correction factors for energy dependence has been shown to be feasible by several studies. C orrection factors from 0.81 1.56 have been suggested for effective energies ranging from 29 62 keV. 73 Lavoie et al. found correction factors for CT energies specifically, ranging from 0.99 1.22 in terms of ion chamber to OSLD response. 72 In this

PAGE 57

57 research, additional work was done in order to appropriately correct for energy depth dependence that can account for both the effects of C T energy spectra and geometry; thus, correction factor s, C E,S were determined and applied to the measured doses This is described in 4.2.2. 4.1.2.3 OSL a pplications in c omputed tomography The OSL N anodots have been used to measure skin dose s in CT. A study on head CT protocols was conducted at the Mayo Clinic (Rochester, MN) where an anthropomorphic phantom was used to assess differences in the dose to the lens of the eye from the application of organ based TCM, global mA reduction, and the use of bismuth eye shields. 77 Another study used OSLDs to measure thyroid doses to pediatric patients in CT. 78 In addition to N anodots Landauer has begun ma king strips of OSL material within an acrylic rod that can be used to measure dose profiles in CT. The OSL strips must be sent back to the manufacturer for reading and interpretation. 76 4.2 OSLD Measurement Methods 4.2.1 Nanodot Reading and Measurement P rotocol T o ensure consistency, reliability, and reproducibility, a reading protocol, as established in the previous work done by Tom Griglock and Lindsey Lavoie was followed 79,72 The reading protocol establishes the following steps: 1. F irst, the background of all d osimeters is read. All dosimeters that have a background reading greater than 1mGy are put aside and are not used for that particular measurement session 2. Second, the measurement session is performed. For each CT protocol tested in each of the cadaveri c subjects, two identical scans were conducted with the same set of dosimeters in place. As shown by Griglock, this results in a percent deviation of 5%, which makes unnecessary further repetitions of each measurement. 79

PAGE 58

58 3. After the dosimeters are expose d, they are read in the Microstar and doses a re recorded. A great advantage of the OSLDs is the ability to perform repeat readings on the same dosimeter following a given exposure; however it has been shown previously that when reading a OSLD three time s, the increase in the accuracy of the measurement is limited to 1.1%. 79 With this in mind and d ue to the high number of dosimeters being read and analyzed in this research each dosimeter is read only once to make the measurement session more efficient. 4. Finally, a fter doses have been read the OSL material from each Nanodot i s displaced from its plastic casing and exposed to UV light for at least 24 hours. If background doses are found to remain over 1 mGy after 24 hours, longer erasure times are util ized to ensure proper erasure. 72 4.2.2 Dosimeter Calibration Prior to performing dosimetry the Microstar reader was calibrated according to the Landau er operations manual (Landauer, Glenwood, IL). The Microstar reader must be calibrated annually, each year u sing a new set of calibration dosimeters which must be obtained directly from the manufacturer. This procedure calibrates the reader to an 80 kVp beam with a half value layer of 2.9 mm Al. Since the x ray beam s used in CT are characterized by generally h igher energies and half value layers, additional calibration factors must be applied to the raw dosime ter reading In order to calculate dose to tissue from the Microstar rea dout, the following equations a re employed: 72 ( 4 1 ) where D raw is the raw dose reading from the Microstar; ( 4 2 ) and;

PAGE 59

59 ( 4 3 ) T he traditional f factor must be applied to this raw dose in order to convert the dose to air to dose to tissue. The mass attenuation coefficients used to calculate the f factor are a function of effective energy. 67 The HVLs measured were used to compute the f factors for all energy and filter combinations used in the clinical CT protocols investigated As mentioned above, corrections for energy and scatter denoted C E,S must be applied to the raw dose reading s C alibration of the OSL dosimeters for the CT beams used in this research, were done so using a 0.6 cc ion chamber to obtain dose in air measurements traceable to NIST standards This small volume ion chamber has been used in the field of dosimetry for many years and is know n for its flat energy response. 14 Previously, Lavoie explored the energy and scatter effects on the OSLD response by placing increasing thicknesses of acrylic between the x ray source and the dosimeter or ion chamber. This procedure was performed both with a general purpose x ray tube, as well as, a s tationary CT x ray tube. 72 While Lavoie fully explored the energy response of the OSLDs, correction factors we re only obtained for the energy and filter combinations that corresponded with the protocols measured in that study. Because new protocols and corresponding energy filter combinations were used in this project appropriate additional corre ction factors were obtained. Correction factors obtained for this project can be broken up into two categories, surface corr ection factors and organ correction factors.

PAGE 60

60 4.2.2.1 Surface dosimeter c alibration T he surface calibration procedure consisted of placing the small volume ion chamber at scanner isocenter on top of one inch of acrylic. These measurement conditions mimic those of the manufacturer calibration (Landauer, Glenwood, IL). These measurements were performed with the scanner operating in service mode so the x ray tube could remain stationary at the top position in the gantry. After an exposure was recorded with the ion c hamber, this was withdrawn and a Nanodot was placed in the exact location of the center of the ion chamber. Four dosimeters were used for e ach energy/filter combination. The average of these measurements along with the average of the ion chamber measure ments, was used to calculate the surface correction factor. Figure 4 3 shows the measurement configuration. The most commonly used energy filter combination for all subjects and all body protocols investigated in this research was 120 kVp with the large b owtie filter. Verification of the surface correction factor for this energy and filter combination was completed on the surface of a cadaveric subject using the CT beam rotating geometry. The ion chamber was placed on the subject with 4 OSL dosimeters su rrounding it, and two exposures were made. The correction factor was applied to each of the four dosimeter readings and comparisons were made with the ion chamber air kerma reading. Section 8.3 shows the results of this procedure. 4.2.2.2 Organ dosimeter c alibration Organ dose correction factors for CT energies were determined utilizing the standard PMMA cylindrical phantom s, used in CTDI measurements (West Physics Consulting, Atlanta, GA) These correction factors therefore account for variations in depth within a homogeneous cylindrical phantom in order to maintain reproducibility, as

PAGE 61

61 well as, taking into account the geometry characterized by a rotating x ray tube. In this w ay, the 0.6 cc ion chamber was placed at the central (i.e., midway through the width) position along the z axis in the phantom, and measurements were conducted The scan parameters a nd position of the phantom in the gantry remain ed constant, with only the energy and filter being changed. To ensure geometr ic consistency in the calibration measurements these were performed with the scanner operating in service mode so that its SYNC function could be utilized. The SYNC function allows the user to synchronize the start angle of the x ray tube in the gantry f or every measurement After all measurements were conducted with the ion chamber, a Nanodot was placed in the center of the phantom and exposed. This process was repeated for each energy/filter combination and each depth in the phantom. The organ dose correction factors were then calculated from these measurements by taking the ratio of the dosimeter measurement to that of the ion chamber measurement. For organ dose measurements for head CT protocols, the dosimeter response was examined utilizing the he ad PMMA phantom, 16 cm in diameter. Two different measurement positions were examined within the phantom, the periphery and center. The mean of the correction factors at the periphery and center was applied to the raw doses measured for the brain and thy roid measurements for all adult head protocols. In the case of body CT protocols, the larger 32 cm diameter PMMA nested phantom was utilized to analyze dosimeter response. T he nested body phantom allows measurements to be made at three different depths, i.e., the center, middle, and periphery. The peripheral position corresponds to a depth of 1 cm and the middle position corresponds to a depth of 9 cm The mean of the results of all three

PAGE 62

62 measurement depths was applied to all organ doses measured for a ll adult body protocols. The measurement setup for the organ dosimeter calibration, specifically adult body CT protocols, is shown in Figure 4 4. 4.3 Measurement Mediums Current organ dose estimates are based upon two methodo logies. One is that of physica l dose measurements with a specific CT scanner and physical phantom. The other methodology for dose estimation involves a Monte Carlo model of the source with corresponding simulations performed on various computational phantoms. These two methods are br iefly described below, as well as, a new methodology involving the use of cadaveric specimens as a measurement medium. 4.3.1 Physical Phantoms A plethora of physical phantoms exist for use with CT dosimetry. As mentioned earlier, two cylindrical PMMA 15 cm long phantoms exist for the measurement of CTDI. In addition, physical phantoms have been constructed that serve to mimic real patient anatomy and corresponding tissue densities. Two commercially available types of anthropomorphic phantoms include the Rando Alderson phantom (Radiology Support Devices Inc., Long Beach, CA) and the CIRS or Computerized Imaging Reference Systems phantoms (Norfolk, VA). A research group at the University of Florida has also developed a set of anthropomorphic phantoms made from a ctual CT image data sets. 80, 81 Figures 4 5 and 4 6 show examples of the CIRS and UF anthropomorphic phantoms. Physical phantoms offer a real advantage in their convenience of use. A quick and meaningful dose comparison for the introduction of a new CT protocol or the optimization of an old one can be completed with the use of these phantoms. However, one potential disadvantage associated with the use of these phantoms is that they are

PAGE 63

63 generally limited to three tissue equivalent materials for bone, so ft tissue, and lung. That being said, physical phantoms can also be used to validate Monte Carlo models and can estimate organ doses for more advanced CT technology, like tube current modulation, which has proved difficult to model. 4.3.2 Computational Phan toms The first generation of computational phantoms was stylized. Stylized phantoms consist of geometric shapes and mathematical surface equations that are combined to resemble a simplified model of human anatomy. While these models were first designed i programs for computing CT doses to patients. 82, 58 The next set of computational phantoms was termed tomographic, owing to their origin of CT datasets. These phantoms are much mor e anatomically realistic, but are very difficult to scale or alter. A third generation of computational phantoms called hybrid phantoms has recently been developed. Human anatomy within hybrid phantoms are made up of NURBS (Non Uniform Rational B Spline) and polygon mesh surfaces. They retain the anatomical realism of tomographic phantoms, and can also be easily modified to model different patient statures and weights. 83, 84, 60, 61 A main contributor to the development of hybrid phantoms is the Bone Im aging & Dosimetry research group at the University of Florida. They have published hybrid phantoms representing the ICRP89 reference newborn, 1 year, 5 year, 10 year, 15 year and adult male and female. 60, 61 Figure 4 7 shows several of these UF hybrid pha ntoms. The evolution of computational phantoms has resulted in many currently accurate representations of patient anatomy. In the application of these phantoms to CT dosimetry, two main stipulations exist. First, as mentioned earlier, most widely used

PAGE 64

64 CT software programs available are based on the earlier, less accurate stylized phantoms. Second, because of the complexity of CT acquisitions, including geometry, filtration, tube current modulation, and new reconstruction algorithms, it is extremely diffi cult to accurately model CT with Monte Carlo simulations. Still, new and improved CT dosimetry software is currently under development in an effort to improve upon Monte Carlo modeling with the use of up to date computational phantoms. 4.3.3 Cadaveric Subject s Cadaveric specimens have only recently been used as a measurement medium for CT organ doses. Dr. Thomas Griglock was the first to make a comprehensive set of organ dose measurements with cadaveric subjects for a variety of CT protocols. His methods of tube placement and dose measurement are used with this project. 79 Using cadavers to closely resemble live patient anatomy requires a review of recent data on the postmortem changes that could affect organ doses in computed tomography. Applications of MD CT technology have been realized in forensic imaging. Over the last ten years, multiple studies have been performed concerning postmortem effects are livor mortis, rigor mortis, and decomposition. Livor mortis is the settling of the blood after death, rigor mortis is the stiffening of joints and skeletal muscles, and decomposition involves the breakdown of matter. Livor mortis is mainly found in the lower portion of the lungs and exhibits a position dependency. A position dependent density change can also occur in live patient scans when proper inspiration is not achieved. Postmortem imaging mimics a scan with patient expiration and partial collapse of the lungs. 85

PAGE 65

65 Hype rattenuation can occur in the major arteries and cardiac chambers due to hemoconcentration associated with livor mortis. 86 However, care must be taken to distinguish between a hyperdense aortic wall as a result of arteriosclerosis. 87 Hyperattenuation has also been observed in the posterior portions of the sinuses around the cranial fossa. In contrast, rigor mortis has been shown to not affect attenuation or shapes of skeletal muscles. Decomposition occurs in stages and the corresponding imaging effects can be classified as early, moderate, or advanced. 86 Early signs of decomposition include cerebral autolysis and the loss of differentiation of gray and white matter of the brain as a result of brain swelling. 87 In early to moderate stages of decompositi on, small amounts of subcutaneous and visceral gas are observed within the abdominal cavity. In the late stages of decomposition, organs begin to collapse, the settling of the brain in the posterior portion of the cranium is observed and liquefaction begi ns. Additionally, there is evidence of insect activity throughout the cadaver. 86 From this information, it can be deduced that early stages of decomposition will not significantly affect organ dose measurement in MDCT. Furthermore, Levy et al. found th at the attenuation of the liver, spleen, kidneys and solid visceral organs has been shown to not exhibit changes until the late stages of decomposition. 86 Postmortem imaging was also examined as an additional teaching module for gross anatomy courses at Wa ke Forest University. A study of 18 embalmed cadavers was conducted with an average time between embalming and scanning being 8 months. Their findings were extremely similar to those discussed above, such as

PAGE 66

66 hyperattenuation of the aortic wall, air fluid levels in the bowel, loss of differentiation between gray and white matter, and fluid in the tracheobronchial tree. 88 To our knowledge, the first time cadaver specimens were used in assessing doses from CT exams was in 2001 to examine absorbed dose differ ences from conventional spiral tomography and spiral CT for an orofacial protocol. The study involved measuring thyroid, lens, parotid gland, and submandibular gland organ doses in a cadaveric head and phantom head with TLDs. 89 In 2008, another study uti lized cadaver heads to examine a neurology protocol and measured lens doses from 16 and 64 slice MDCT scanners. It needs to be noted that this study only utilized cadaveric heads that were frozen. In addition, the brains were previously removed and the h ead cavity was filled with gauze. 90 Hounsfield units and corresponding attenuation differences for the cadavers and 5 patients. Table 4 1 shows the percent differences between the c adavers and live patients. In conclusion, due to the small attenuation differences observed, cadavers can be used as a representative of a live patient for measuring organ doses in CT. 79 Finally, the changes observed in postmortem anatomy will not likely affect the organs of interest in this project.

PAGE 67

67 Figure 4 1 Landauer Nanodot from the InLight Microstar Dosimetry System (Landauer Glenwood, IL ) Figure 4 2 Simplified optically stimulated luminescent ( OSL ) reader s chematic. Adapted from Yukihara EG, McKeever SWS. Optically Stimulated Luminescence (OSL) dosimetry in medicine. Phys. Med. Biol. 2008; 53: R351 379.

PAGE 68

68 Figure 4 3. Surface dosimeter calibration setup. A) Ion chamber B) Nanodot Figure 4 4. Organ dosimeter calibration setup for body protocols.

PAGE 69

69 Figure 4 5. CIRS anthropomorphic phantoms (Norfolk, VA). Figure 4 6 UF anthropomorphic phantoms 81

PAGE 70

70 Figure 4 7 UF library of ICRP89 reference hybrid computational p hantoms 61 Table 4 1 Attenuation differences for a patient versus a cadaver 79 Tissue Type Hounsfield Units % Difference versus Patient Patient Cadaver Lung 772 750 9.6% Fat 102 75 3.0% Muscle 50.6 55 0.4% Bone 555 550 0.3% Kidney 42.6 50 0.7% Liver 56 45 1.0%

PAGE 71

71 CHAPTER 5 CHARACTERIZATION OF BEAMS IN ULTRA HELICAL ACQUISITION MODE A full characterization of the Aquilion ONE (Toshiba America Medical Systems, Tustin, CA) was done previously by Lavoie for its 64 slice and full 320 slice volumetric modes. 72 With the introduction of the ultra hel ical scan mode, certain beam characteristics must be revisited in order to understand the implications of this type of acquisition on clinical scanning. Properties such as beam quality, beam width, and over ranging were examined and are described in the f ollowing sections. 5.1 Half Value Layer An important characteristic of an x ray beam which determines how radiation is absorbed a nd distributed within a patient is the x ray beam quality. As such, the half value layer (HVL) is the measurement used to quantif y beam quality. The HVL is the thickness of a given material required to reduce the intensity of a photon beam to half of its initial value. For beams in the diagnostic energy range, it is usually described in terms of millimeters of aluminum equivalent. Conceptually, knowledge of the HVL provides a qualitative measure of the penetrability or hardness of a photon beam. The HVL also provides a qualitative measure of the effective energy of the beam, a quite important parameter in CT because of the depend ence of the Hounsfield Unit with beam energy. The HVL also provides a way to compare higher and lower HVL beams in terms of increased skin dose. This is because when a beam is less penetrating (lower HVL), it is characterized by a higher content of lower energy photons. Lower energy photons are more likely to be absorbed by the surface of a patient and less likely to contribute to image formation.

PAGE 72

72 For this study, the first and second HVLs of the beams were measured. The second HVL is the amount of material needed to reduce the beam to 25 percent of its initial intensity. This concept only has meaning for polyenergetic beams because of beam hardening. Clearly, the second half value layer of a polyenergetic beam is greater than the first half value layer. The homogeneity coefficient was also calculated for this research, and is the ratio of the first half value layer to the second half value layer. Because of the higher nominal tube voltages and the use of flat and bow tie filters, HVLs measured fo r CT beams are generally higher than those from beams used in general radiographic or fluoroscopic systems. HVLs were measured with a R10X6 6 ion chamber (Radcal, Monrovia, CA) placed at isocenter. In order to maintain good geometry, a lead aperture was placed directly over the parked x ray tube at the bottom of the gantry. Although devices have been developed to measure the HVL of the rotating beam in CT, a simpler method is to lock the x ray tube in a preselected position and conduct a standard good g eometry HVL measurement. By placing the ion chamber at isocenter, the measurement will be reproducible and will only measure the primary beam as opposed to sca tter from the added aluminum. 5.2 Beam Width Beam widths were measured for not only the new ultra helical detector configurations, but for all configurations utilized in this research. The beam widths for the following detector configurations were measured: 0.5 mm x 32, 0.5 mm x 64, 0.5 mm x 80, 0.5 mm x 100, 0.5 mm x 160, and 0.5 mm x 320. The nomin al beam widths for these detector configurations are defined at isocenter as 16, 32, 40, 50, 80, and 160

PAGE 73

73 mm. The actual beam widths are known to be slightly larger due to the effect of over beaming, described in Section 2.1.2 14 Beam widths were measured on the Aquilion ONE (AQ1) with the use of a large computed radiography (CR) photoluminescent imaging plate placed at isocenter. The x ray tube was locked in position with the scanner working in service mode for a stationary exposure with the tube placed at the top of the gantry. A copper filter was placed over the x ray tube in order to mimic the typical radiographic exposures (approximately 1 mGy) received by the plate. Exposures were made at the configurations mentioned above and the CR cassette was p rocessed via an Agfa CR 85 digitizer (AGFA, Teterboro, NJ) using a flat field algorithm. An example of the images produced is shown in Figure 5 1 The data was analyzed with iSite software (Philips Healthcare, Andover, MA), with which the width of the beam was measured from the CR image. The ratio of the nominal beam width to the actual beam width is a measure of geometric efficiency. 5.3 Over R anging As mentioned earlier, over ranging refers to the additional rotations or scanning length needed for reconstruction of a helical acquisition. As more data acquisition rows have been added to detector arrays in MDCT systems and the beams continue to get wider, geometric efficiency is increased, effectively lowering patient dose. However, because the beam width is increasing, the additional rotations pre and post scanning will certainly increase the irradiated regions substantially. A fundamental part of a clinical CT protocol is the standardization of the anatomical landmarks where the scan begins and e nds. In addition to anatomical and clinical reasons, one of the criteria for such is to avoid exposing radiosensitive organs. Thus, the over ranging may result in

PAGE 74

74 irradiation of radiosensitive organs that otherwise would not be exposed to the primary bea m. 24 Although the effect is quite evident with 64 slice scanners, it is assumed that the effect will worsen with 80, 100, and 160 channels used for data acquisition in ultra helical mode. To address this issue active collimation has been installed on all newer models of MDCT scanners including the AQ1. The additional length of the scan caused by over ranging, will be referred to as over ranging length in this document. There are two methods to calculate over ranging length. The discrepancy between the two methods is thought to be a result of a proprietary, the reason for the difference can only be investigated though not necessarily described; both methods were employe d in order to examine the effect. The first is a simple estimate, where the total imaging length of the scan is calculated by dividing the DLP (mGy*cm) by the CTDI Vol (mGy), from the dose report from each exam. The planned scan range is the range select ed by the technologist prior to image acquisition according to the standardized protocol. The over ranging length is then calculated by subtracting the planned scan range from the total imaging length. The second method to calculate over ranging lengt h is described i n detail in AAPM Report No. 111. 14 The total scanning length, L, is quantified by the following equations: ( 5 1 ) where

PAGE 75

75 ( 5 2 ) ( 5 3 ) In the equations above, v is the table speed and t The table speed is the ratio of the table increment per rotation, b, and the rotation period, The table increment is calculated by the multiplication of the pitch factor ( p ) and the nominal beam width ( n x T ). Once the total scanning length is calculated, the over ranging length is found by subtracting the planned scan range from the total scanning length. Both methods were utilized to calculate over ranging length for the specific CT protocols studied in this project. This effect was examined for CT protocols which have radiosensitive organs just outside the planned scan range. In this way, it is possible to compare doses with the over ranging length calculations to determine the clinical implications of this issue. T he CT protocols used to examine this effect include the standard head protocol, the chest protocol, and the abdomen protocol. For the standard head protocol, the thyroid lies just outside the predetermined scan length and hence, was used to quantify the o ver ranging effect. The small intestine and colon are organs outside the planned scan range for the chest protocol. For the abdomen protocol, the upper lungs and uterus are the organs that will be examined.

PAGE 76

76 Figure 5 1. C omputed radiography image of 0.5 x 160 mm beam w idth.

PAGE 77

77 CHAPTER 6 DIRECT ORGAN DOSE MEASUREMENT METHODOLOGY The measurement methodology explained here is based on the dissertation of Thomas Griglock. 79 It will be summarized in this section, as well as any changes made in an effort t o broaden the scope of the measurements and increase the accuracy of measurements. The in situ methodology encompasses two main aspects: the tube placement and the dosimeter placement. 6.1 Tube Placement The first step needed to perform actual organ dose me asurements within cadavers is tube placement. In order to accurately describe that process the tube placement system must first be explained. The tube placement system allows for consistent and reproducible external access to internal organs. The tubes consist of clear PVC (polyvinyl chloride) material with an outer diameter of 19 mm and an inner diameter of 16 mm. The 16 mm inner diameter allows for the dosimeter to easily slide are placed within organs and are not moved until all measurements have been completed and the cadaver is ready to be returned. In an effort to keep the inside of the tube uncontaminated, the bottom end is sealed prior to being placed within t he body. The next aspect of the tube placement system is the dosimeter holders, which consist of 2 much smaller tubes or straws. The innermost straw is used to create holders for the dosimeters with a sealer, while the outermost straw is simply used as t o reinforce the inner straw and provide structural stability. Depending on the organ, more or less dosimeter holders will be made with the inner straw. Figure 6 1 is a depiction of the tube placement system.

PAGE 78

78 The organs in which dose measurements would be made were determined based on their radiosensitivity, considering stochastic and deterministic effects, as well as, the feasibility of measurement. The organs chosen include the brain, lens, thyroid, breasts, lungs, liver, stomach, colon, small intest ine, ovary, uterus, and the skin. It must be noted that a major radiosensitive organ is not included in this list, the bone marrow. Obtaining dose information within the skeletal structure of a cadaveric subject is not feasible with the methodology of th is project This is an instance where computational models offer a real advantage. Figure 6 2 is a schematic of the organ tube placement. A board certified radiologist, Sharatchandra Bidari MD, performed the tube placement for all cadavers. Prior to t he start of tube placement, a full body scan of the subject was completed. This allowed visualization of all of the organs and identification of proper access points. A biopsy grid was placed on the surface of the subject in order to accurately identify planned locations of tube insertion. During tube placement, scans were repeated multiple times to ensure the tube was placed correctly inside the organ of interest. After all the tubes were placed, a final verification scan was completed to double check the position. Finally, the tubes were marked so that any deviation in position over time could be observed. 6.2 Organ Dose Measurement Once tube placement was completed for each subject, dosimetry could be performed. Before the start of this project, a re evaluation of dosimeter placement from the previous methodology was conducted. 79 It was concluded that for larger organs, increasing the number of dosimeters could improve the accuracy of the model. For those organs that are paired, it was decided to measure both the left and right side, even though Griglock showed they may only differ by 5%. 79 The lungs and the breasts

PAGE 79

79 are included in this discussion. Due to these organs being fairly large and rad iosensitive, it is considered advantageous to have more dose information for them. For example, there were 4 tubes placed in the lungs, 1 upper tube and 1 lower tube for both the right and left lungs. Dosimeter placement was planned so that the dosimeter s were sure to remain in the organ of interest and to cover as much area as possible. Hence, 2 dosimeters were placed in each lung tube. Figure 6 3 shows tube and dosimeter placement for the right breast, left lobe of the liver, and the stomach. The tu be placement coupled with the dosimeter placement ensured accuracy, and that a wide area of anatomical coverage was achieved. Table 6 1 shows the tube and dosimeter logistics for each organ. The number of dosimeters used for each protocol was determined by which organs could be within the planned scan range. The maximum number of dosimeters used for any single protocol was 59, which was for the trauma protocol 6.2.1 Surface Measurements The number of surface dosimeters was also increased from the previous methodology. 79 For every body scan, 15 dosimeters were used to assess skin dose. The surface dosimeters were prepared in strips, containing 5 dosimeters per strip placed 5 cm apart. T hree strips were used for every scan, and placed 10 cm apart along the z axis of the patient. It is important to keep the surface measurements in the primary beam for all scans in order to maintain consistency. Hence, for scan ranges that were much short er than a CAP, the z axis distance between the strips was shortened to 5 cm. Figure 6 4 shows how the surface dosimeter strips were placed for every subject.

PAGE 80

80 6.2.2 The imaging of pregnant patients is medically necessary in certain clin ical situations. When CT exams of pregnant patients are needed, careful attention must be paid to the radiation dose absorbed by the fetus. In order to answer if the benefit outweighs the risk, it is necessary to know the risks. The risks to the fetus as a result of radiation exposure during the first trimester include prenatal death, congenital malformations, mental retardation, cognitive impairment, microcephaly and the development of childhood cancer. These biological effects vary with gestational age and by the magnitude of the dose absorbed by the fetus. 91 The NCRP, ICRP, ACR (American College of Radiology) and the ACOG (American Congress of Obstetricians and Gynecologists) are a few organizations that address fetal effects at diagnostic dose levels There are a few variations, but the general consensus is that with a dose below 50 mGy, the risks to the fetus are negligible, and that fetal doses below 100 150 mGy, warrant no further action. 91, 92, 93, 94 The research presented in this document will serve to answer a two fold question: first, what values of fetal dose can be expected from the CT imaging of pregnant patients; and second, whether or not such values fall within the levels of negligible risk. Fetal doses, thus far, have only been quantif ied using either physical anthropomorphic phantoms or computational phantoms, but never have been directly measured within a human body due to obvious practical limitations. 95, 96 97, 98 As with many studies of this nature, the dose to the uterus has bee n used as a surrogate for the subjects to measure and estimate first trimester fetal doses. An example of dosimeter placement for this measurement is shown in Figure 6 5 The three main CT protocols of

PAGE 81

81 relevancy for imaging pregnant patients, and examined in this research, are the chest abdomen pelvis, trauma, and pulmonary embolism protocols. The doses that result from these protocols were examined for the 0.5 mm x 64 detector configuration, as this is the most common configuration available currently. These protocols are described at length in Section 6.3.2 below. 6.2.3 Head Measurements Articles within the public media have drawn significant attention to the unfortunate ca ses of radiation burns experienced by several Cedars Sinai patients in June 2009, who were exposed to excess radiation during brain perfusion head CT scans. 5 Standard head and CTA head protocols have been some of the most common procedures performed durin g the past ten years, owing to the efficacy of CT scanning to diagnose strokes and bleeding within the brain. In addition, CT brain perfusion scans are now being performed on both narrow and wide beam scanners, such as the Aquilion ONE (Toshiba America Me dical Systems, Tustin, CA), which can image the entire brain in a single rotation. Doses from volumetric acquisitions of the brain have been shown to be lower than those from conventional 64 slice scanners. 72 The organ dose measurements in this work aim t o better quantify the dose differences between detector configurations on the same scanner for various head CT protocols. The fifth cadaver acquired for this study, was selected to be a torso with a cephalus in order to include dose measurements for common head CT protocols in the project. The sixth and seventh cadavers were full body subjects, which also allowed for head dose measurements to be conducted. Based on the relative radiosensitiv ity of organs within the head, it was decided that the only inter nal organ dose measurements required were for the brain. Thus, two tubes were used for the brain dose

PAGE 82

82 measurements. An international fellow of neurosurgery, Dr. Maria Peris Celda performed a craniotomy to insert the tubes. The first tube was inserted in to the frontal lobe and extended to the thalamus. The second tube was inserted at the parieto occipital fissure and extended centrally 8 cm. This allowed for doses to be measured at varying depths within the brain. A total of 12 dosimeters were used for each head study. They consisted of 2 in the thyroid, 3 on the surface, 2 placed on the eyes, and 5 dosimeters placed in the brain tubes. There were 2 dosimeters placed in the frontal tube, separated by 4 cm. Three dosimeters were placed in the posterior tube, 3 cm apart. Figure 6 6 shows an example of dosimeter placement within the head. 6.3 CT Protocols With this research, o rgan doses were measured for the fo llowing CT protocols: standard head, computed tomography angiography ( CTA ) head, brain perfusion, chest abdomen pelvis, chest, abdomen, pelvis, three phase liver, pulmonary embolism (PE) and trauma. All protocols were selected based on how common they were performed or if they had the potential to result in higher than average doses, such as the three phase liver, trauma, and brain perfusion protocols. These protocols are further distinguished as head or body protocols. However, the trauma protocol does consist of both head and body scans, in order to completely assess the do se to a trauma patient. All of the protocols described here have been optimized and approved by the RPC or Radiology Practice Committee at SUF. The main job of this committee is to maintain consistency in the way that CT scans are performed in the depar tment. With

PAGE 83

83 the efforts of the RPC, all protocols have been standardized, which ensures uniformity across all scanners in this institution. When organ doses are reported for these various protocols, it must be understood that only those organs which are w ithin the primary range of each protocol are reported. Organs that are completely outside of the primary range are assumed to only be exposed to scattered radiation, which results in minimal dose. This is further described in Section 9.3. Organs that ar e exposed as a result of over ranging are separately acknowledged in Section 9.1.1. Additionally, it must be identified that some of the protocols described here entail the use of contrast to enhance pathology. These include the three phase liver, pulmonar y embolism, and trauma protocols. The use of contrast was not able to be simulated in these protocols due to the cessation of blood flow within the subjects being studied. Thus, the organ doses reported for these protocols may in fact differ slightly fro m doses absorbed in live patients undergoing these exams. However, the exposure conditions for the protocols mentioned here were able to simulated, i.e. the amount of exposure time needed for bolus triggering. Therefore, the only differences in dose betw een these protocols performed on live patients versus deceased, is the increased attenuation caused by the contrast within the vascular system. In conclusion, the assumption is made that the impact of these differences on the reported organ doses will be minimal. The following sections comprehensively describe all protocols investigated in this study

PAGE 84

84 6.3.1 Head Protocols 6.3.1.1 Head without c ontrast A standard head protocol at Shands UF (SUF) includes a helical head scan with no use of contrast. The scan begins at the base of the skull and ends at the skull ve rtex, as displayed in Figure 6 7 Indications for this exam include: headache, intracranial hemorrhage, hyrdocephalus, brain injury, seizure, and head trauma. Standard head protocols at SUF are performed at 120 kVp, 270 mA, 0.75 s rotation time, a pitch of 0.66, a range of 160 mm, and with a small or medium filter. 6.3.1.2 CTA h ead A CTA head protocol, as performed at SUF, starts with a standard head scan without contrast. Contrast is then administered and the bolus is triggered at the hyoid. Upon bolus triggering, a helical scan is immediately acquired from the level of the C3 vertebra to the skull vertex. A pictor i al display of this scan acquisition is shown in Figure 6 8 A delayed pos t contrast acquisition is the last phase of this protocol. This acquisition is in accordance with the standard head protocol described above. Clinical indications for this protocol include: subarachnoid hemorrhage, vascular bypass surgery, aneurysm, vasc ular abnormality, vertebrobasilar insufficiency, dissection, and stenosis. Both the standard head and CTA head protocols were evaluated at the following detector configurations: 0.5 mm x 32 and 0.5 mm x 64 (helical mode), 0.5 mm x 80, 0.5 mm x 100, 0.5 mm x 160 (ultra helical mode), and 0.5 mm x 320 (volumetric mode). The 0.5 mm x 320 detector configuration can only be operated in volumetric acquisition mode, with a rotating gantry and stationary table.

PAGE 85

85 The standard head protocol is currently performed a t SUF with a detector configuration of 0.5 mm x 32. The optimized CTA head protocol consists of a standard head scan without contrast at a detector configuration of 0.5 mm x 32, followed by a CTA helical acquisition at a detector configuration of 0.5 mm x 64, ending with a delayed scan, which is completed at a detector configuration of 0.5 mm x 32. This the CTA head protocol at alternate detector configurations, it must be unde rstood that identical detector configurations were used for all helical phases of the protocol including, the initial head without contrast, the CTA phase, and the delayed head without contrast. The detector configuration utilized is indicated in the tab les in 8.3.1. The helical portion of the CTA head protocol is completed with similar scan parameters as the standard head scan, with the exception of the tube current, which is increased from 270 mA to 400 mA. 6.3.1.3 Brain p erfusion The final head proto col examined in this research included the brain perfusion CT protocol. Brain perfusion scans can deliver higher organ doses than the head scan protocols mentioned in the section above. Therefore, high priority was given to the measurement of organ doses from this protocol, specifically dose to the lens of the eye In order to assess the dose to the lens from the brain perfusion protocol and to evaluate potential dose reduction techniques, the lens dose was measured with and without the use of a bismuth shield for 2 of the 3 cephalic subjects. The clinical indications for a brain perfusion protocol include vasospasm and acute stroke. The use of CT for the evaluation of these clinical indications can result in rapid diagnosis and better treatment option s for patients. By utilizing the full detector

PAGE 86

86 array of the Aquilion ONE (Toshiba America Medical Systems, Tustin, CA), the entire brain can be imaged in one rotation. The brain perfusion protocol begins with a standard head scan without contrast. This acquisition is performed in order to determine if brain hemorrhages are present. It is followed by the perfusion acquisition series, which are performed in the dynamic volume mode of the 320 slice scanner. The scan range is 160 mm and is denoted by the same anatomical i ndicators as shown in Figure 6 7 The dynamic volume mode consists of a series of timed delayed volumetric acquisitions obtained with the table remaining stationary; this results in repeated exposures over the same region of anatomy. Thi s exposure sequence is shown in detail in Figure 6 9 The final scan for this protocol is a delayed standard head acquisition. Both of the head scans performed immediately before and immediately following the perfusion acquisition series are performed wi th the Shands default protocol, which is acquired at a detector configuration of 0.5 mm x 32, as mentioned above. In some cases, such as those when patient motion occurs during the perfusion portion of the scan, a CTA head acquisition is performed after the perfusion acquisition series and before the delayed head scan. Because this represents a special case, this CTA head acquisition was not included in the measurements conducted for this research. 6.3.2 Body Protocols 6.3.2.1 Chest abdomen p elvis The chest abdomen pelvis (CAP) protocol scans the entire torso, starting at the thoracic inlet and ending at the lesser trochanter. The indications for a CAP protocol include, but are not limited to, a fever of unknown etiology, search for primary

PAGE 87

87 maligna ncy, colorectal cancer, lymphoma, and workup of metastatic disease. All adult CAP exams are performed at 120 kVp, a tube rotation time of 0.5 s, with the large filter, a pitch range of 0.81 0.87, and tube current modulation activated. 6.3.2.2 Chest The ch est protocol calls for images from the thoracic inlet to the top of the kidneys. Due to the asymmetrical nature of the kidneys, the scan is usually ended at the top of the right kidney, since it is inferior to the left kidney. A chest CT can be requested if abnormalities are discovered on a chest radiograph or if occult thoracic pathology is suspected clinically. The protocol can also be requested for staging and follow up of lung cancer or other malignancies. Thoracic vascular abnormalities or anomalie s can be diagnosed with a chest CT. Finally, a chest protocol can be used for the evaluation and follow up of pulmonary parenchymal and airway disease. All adult chest protocols are performed at a nominal tube voltage of 120 kVp, a tube rotation time of 0.5 s, and with a large filter. The pitch factor ranges from 0.99 to 1.48, higher than most body protocols so that a faster scan time can be accomplished and motion artifacts are minimized. 6.3.2.3 Abdomen An abdomen protocol starts at the dome of the d iaphragm and ends at the iliac crest. Indications for this exam include the following: abdominal pain, abdominal hernia, weight loss, nausea and vomiting, fever, non complicated pancreatitis, evaluation of renal masses, and detection of complications afte r abdominal surgery. The scan parameters for the abdomen protocol mimic those of the CAP protocol.

PAGE 88

88 6.3.2.4 Pelvis A pelvis scan begins at the iliac crest and ends at the lesser trochanter. Clinical indications for a pelvic CT include: pelvic pain, ovarian cyst, perirectal abscess, pelvic mass, pelvic inflammatory disease, and detection of complications after pelvic surgery. Scan parameters for the adult pelvis are identical to those of the CAP protocol. 6.3.2.5 Three phase l iver The three phase l iver protocol has the same scan range as the abdomen protocol, starting at the dome of the diaphragm and ending at the iliac crest. However, it entails three different helical acquisitions over the same anatomical area. This protocol requires the use of contrast. When contrast is injected, the bolus is tracked at the superior end of the scan and is triggered on the aorta. The first helical scan is acquired immediately when the bolus is triggered and represents the arterial phase of this protocol. The v enous phase is initiated after a 70 second delay. The third and final phase is known as the delayed phase, and is performed 3 minutes following the venous acquisition. The clinical indications for a three phase liver scan include a possible liver mass, p re transplantation evaluation, cirrhosis, portal vein invasion, neuroendocrine tumor, hemangioma, cholangiocarcinoma, and hepatocellular carcinoma. All three phases of this exam are performed at 120 kVp, 0.5 s tube rotation time, a pitch between 0.81 and 0.87, and with the large body filter. Tube current modulation is also employed for all helical phases. 6.3.2.6 Pulmonary e mbolism A pulmonary embolism is the one of the most frequent CT prot ocols used on pregnant patients. This protocol is used for t he primary diagnosis of pulmonary embolism in patients that have presented with clinical findings suggestive of the

PAGE 89

89 pulmonary embolism or when a rapid diagnosis is needed. The scan range is similar to a chest CT, starting at the base of the adrenals and e nding just above the apex of the lung. The use of contrast is essential for this exam. The contrast bolus is tracked and triggered on the pulmonary artery, at which time the helical acquisition commences. The scan parameters for the helical acquisition are almost identical to those used in a chest protocol. The pitch is the only parameter that differs from the chest protocol, ranging from 0.81 to 0.87 similar to the other body protocols examined here. 6.3.2.7 Trauma A trauma protocol is commonly u sed in emergency situations to evaluate internal complications in a timely manner. A full trauma protocol requires scans of the head, neck, and torso. The protocol begins with a standard head scan, and is followed by a scan of the cervical spine. The he ad scan is completed at 120 kVp, 290 mA, 0.75 s rotation time, and a pitch between 0.64 and 0.87. With the c spine portion of the protocol, tube current modulation is active and the scan is completed at 135 kVp with the same tube rotation time and pitch a s the head scan. The small and medium filters are most commonly used for portion of the exam. The body portion of the trauma protocol requires the use of contrast, for which the contrast bolus is tracked at the level of the aortic arch and triggered on th e descending aorta. At the time of triggering, an arterial CAP scan is initiated. A venous abdomen pelvis (AP) scan follows after a 70 second delay. The abdomen pelvis scan starts at the dome of the diaphragm and ends at the lesser trochanter of the femu r. In some cases, an additional delayed scan will be performed to image the urinary tract. The dose measurements reported for this protocol do not include the urinary tract

PAGE 90

90 delayed phase. The scan parameters for the CAP and AP scans are identical to tho se listed in 6.3.2.1. This exam is another protocol that is used as an important triage tool for pregnant patients that have experienced trauma, such as a car accident. In some trauma situations where a patient is known to be pregnant, steps can be ta ken to reduce the fetal exposure. Hence, the fetal dose estimate reported for this protocol is a conservative estimate of dose. It must also be noted that the fourth subject was a torso with no cephalus. In order to simulate the exposure from this protoc ol a head phantom was aligned with the subject. However, due to the obvious differences in media, the head and neck organ doses were not reported for the phantom.

PAGE 91

91 Figure 6 1. Tube placement system. Figure 6 2. Tube placement schematic.

PAGE 92

92 Figure 6 3 Dosimeter p lacement in the right breast, left lobe of the liver, and stomach Figure 6 4. Placement of skin dosimeter strips for a chest protocol.

PAGE 93

93 Figure 6 5. Table 6 1. Tube and dosimeter locations. Organ Location # of Tubes # of Dosimeters Brain 2 5 Lens 0 2 Head Surface 0 3 Thyroid 0 2 Breast triple R 1 3 Breast double R 1 2 Breast triple L 1 3 Breast double L 1 2 Upper Lung double R 1 2 Lower Lung double R 1 2 Upper Lung double L 1 2 Lower Lung double L 1 2 Upper Right Lobe of Liver double and single left lobe 1 3 Lower Right Lobe of Liver single 1 2 Stomach 1 2 Small Intestine 1 2 Ascending Colon R 1 1 Descending Colon L 1 1 Ovary 1 2 Uterus 1 1 Surface Strips 0 15 TOTAL 16 59

PAGE 94

94 Figure 6 6 Dosimeter placement for organ dose measurements in the head. Figure 6 7 Scan range for standard head protocol and brain perfusion protocol. Figure 6 8 Scan range for CTA head protocol.

PAGE 95

95 Figure 6 9 Acquisition sequence for brain perfusion protocol.

PAGE 96

96 CHAPTER 7 SIZE PARAMETER MEASUREMENT 7.1 Investigation of Size Parameters An ideal size parameter to characterize body habitus in relation to computed tomography scans should be one that is easy to acquire, can be obtained at the time of scanning, and accurately describes the distribution of size along the body. Height and weig ht were excluded from this study as they offer only basic information on the body build of a patient. In addition, obtaining an accurate weight of a patient at the time of the scan can prove to be challenging. Body mass index or BMI was used solely as a d escriptor of the subjects in this study. Patient perimeter is a parameter that is both relevant to size and has been shown to correlate with dose. 62 However, patient perimeter cannot be easily measured at the scanner console or with standard image proces sing software used in radiologist reading rooms. For this reason, patient perimeter was excluded as a size parameter in this research. The main size parameter explored in this research is the diameter of a circle that has an identical area as the pati ent cross section. This size parameter has been called various names in the literature, such as mean body diameter, equivalent diameter and effective diameter. 99, 100 The term used in this research is the effective diameter, a quantity that is commensura te with that described in AAPM Report No. 204. 52 7.2 Methods of Measuring Size Parameters The effective diameter is based on measurements which can be completed on CT images. From a geometric relationship, the effective diameter can be calculated with the m easurement of the anteroposterior and lateral dimensions, as shown in Equation 7 1.

PAGE 97

97 ( 7 1 ) The effective diameter offers many advantages in being an easily obtained size parameter. However, the quantity also has some ambiguousness surrounding its measurement. AP and lateral dimensions could be measured for any slice within an image set and can also be measured at different locations within a single image. In order to examine the relationship between organ doses and the effective diameter, the effective diameter was measured in two different ways. First, it was measured at the central slice of every scan for all protocols and subjects, as shown in Figure 7 1. Second, it was measured at the central slice of every organ within the image set for all protocols and subjects. Figure 7 2 shows an example of the AP and lateral dimension measurements f or the liver, colon, and lungs. For both methods, the AP and lateral dimensions were measured at the most central location on the patient cross section. An example of the measurement of these two dimensions is shown in Figure 7 1. The CAP protocol was u sed to evaluate which method provided the best correlation with organ dose, due to the inclusion of almost all organs of interest and its clinical relevance. The results of these two methods are compared in Section 8.6.

PAGE 98

98 Figure 7 1. Measurement of the antero posterior ( AP ) and lateral dimensions within the central image for calculation of effective diameter. Figure 7 2. Measurement of the AP and lateral dimensions within the central slice of the following organs: A) Liver B) Colon C) Lungs.

PAGE 99

99 CHAPTER 8 RE SULTS 8.1 Acquisition of Subjects One of the most influential parameters of dose to a patient in CT is body size. For this reason, organ dose measurements were performed on subjects with a range of BMIs or body habitus. In total, 8 adult female cadaveric sub jects were obtained for this research. According to the World Health Organization (WHO), BMI classifications are as follows: 101 Underweight: < 18.5 Normal: 18.5 > 24.99 Overweight: 25 > 29.99 Obese: 30 > 39.99 Extreme Obesity: > 40 The distribution of BMIs of the subjects spanned a wide range, from 16.6 to 43.9, as shown in Table 8 1. BMIs in the overweight or obese classification represent 80.4 % of adult American females of the ages 30 100. 101 Hence, organ dose measurements in the overweight or obe se class ification will represent a large majority of the population. Throughout the rest of this document, the subjects will be described by the numbers listed in Table 8 1. The number refers to order of acquisition of the subjects. For all 8 subjects, organ doses were measured for five of the body CT protocols described in Section 6.3 including a CAP, individual chest, abdomen, and pelvis protocols and a three phase liver protocol. Helical and ultra helical acquisitions were studied with subjects 4, 5 6, and 7 for 7 protocols total. Those 7 protocols were inclusive of the five just mentioned, as well as the pulmonary embolism and trauma protocols. Organ dose measurements for the head protocols were able to be completed with subjects 5, 6, and 7.

PAGE 100

100 8.2 Ch aracterization of Beams in Ultra Helical Acquisition Mode 8.2.1 Half Value Layer The results of the first and second HVL measurements along with the corresponding homogeneity coefficients are listed in tabl e 8 2 The measurement details are described in Section 5.2. The values shown are commensurate with those reported elsewhere in the literatu re. Geleijns et al. measured HVLs of 6.4 and 7.1 mm Al for 120 kVp beams with the small and large bowtie filters in place. 47 8.2.2 Beam Width The measured beam widths and geometric efficiencies for all detector configurations utilized in this project are listed in Table 8 3 For the 0.5 x 320 detector configuration, a beam width of 176 mm was measured, as described in Section 5.3. Lavoie also measured the actua l beam width for this detector configuration, and the same result was reported 72 As expected, the dose efficiency or geometric efficiency increases with beam width. The geometric efficiency ranged from 64 to 91 percent for all the detector configuration s studied. 8.2.3 Over R anging The calculation of over ranging lengths and r otations are shown in Tables 8 4, 8 5, and 8 6 for the head, chest, and abdomen protocols, respectively. These values were calculated by the methods discussed in Section 5.3. The method utilizing AAPM Report No. 111, is denoted TG 111 and refers to the first set of over ranging values. 14 The over ranging lengths calculated utilizing the CTDI Vol and DLP values are listed second. The same scan parameters were used for all subjects for the head protocol, which is the case for all adult heads performed clinically. For the TG 111 method, the

PAGE 101

101 over ranging lengths are 2.62 cm, 4.77 cm, 5.42 cm, 8.36 cm, and 13.65 cm when the number of detector channels used is 32, 64, 80, 100, and 160 respectively. When the DLP is utilized to calculate the over ranging length, the results are 1.78 cm, 2.92 cm, 4.56 cm, 7.11 cm, and 11.09 cm for 32, 64, 80, 100, and 160 active detector channels respectively The CTDI/DLP method results in smaller val ues for all over ranging lengths. The doses corresponding to these over ranging lengths are shown in Figures 8 1, 8 2, and 8 3. 8.3 OSLD Correction Factors The methodology for measuring and computing organ doses is described in detail in Section 4.1.2.4. Eq uations 4 1, 4 2, and 4 3 are utilized to calculate the organ doses measured in this research. Table 8 7 presents the f factors calculated for all energy and filter combinations utilized on the CT scanner investigated in this project. The f factor s used for this project are 1. 06 and 1.07, in accordance with those reported by Lavoie and AAPM Report 96. 72 Tables 8 8, 8 9, and 8 10 compile the energy and scatter correction factors to be used for dosimeters utilized to measure surface doses, as well as those used for internal organ dose measurements, both for head and body CT protocols. The energy and scatter correction factors are presented as the ratios of the dosimeter readings to the ion chamber readings. The last row in each of the tables is the mean c orrection factor for all depths measured within the CTDI phantoms. This is the value that is used as C E,S in equation 4 1, following the same methodology as Lavoie and Griglock. 72, 79 The energy correction factors ranged from 0.83 to 0.92, with the higher energies requiring a greater correction to be made. This result is expected as the reader is initially calibrated to an 80 kVp beam, thus greater departures from this energy will

PAGE 102

102 command a greater correction. For body protocols, the only energy and filt er combination utilized for scanning is 120 kVp, large filter. To correct the skin measurements for body protocols, a correction factor of 0.85 must be applied. This factor of 0.84 for use of the dosimeters with a 120 kVp beam. Verification of this correction factor was conducted on the surface of a cadaveric subject with a rotating x ray tube. The results of this verification are displayed in Table 8 11. Once the correct found to be within 5 % of that measured with the ion chamber. The average of the corrected dosimeter doses was within 1 % of the dose obtained by using the ion chamber. The me an values of C E,S for the head and body phantoms are comparable. Again, the largest correction is required for the 135 kVp beams. The values shown in the tables below are also commensurate with those measured by Lavoie et al. 74 In the body phantom, when the 120 kVp and the small filter are selected, Lavoie et al. reports a correction factor of 0.93, while the correction factor reported here is 0.92. 74 The same difference is observed for the 135 kVp beam, when compared with that of Lavoie et al. 74 Cor rection factors for all energy and filter combinations utilized in the adult CT protocols investigated in this project were obtained with this method. All organ and surface doses reported in this dissertation were appropriately corrected based on the ener gy response of the dosimeters and measurement conditions. 8.4 Organ Doses The organ dose results for the methods described in Chapter 6 are presented in this chapter. For all organ doses measured, the maximum dose, average dose, and

PAGE 103

103 standard deviation are re ported. The majority of the organ doses measured and reported here are those that are in the primary scan range of the protocols examined. Although for some subjects, the same organs are not always exposed due to anatomical variations. In order to acco unt for this variability, some organs were divided into superior and inferior sections. When this division in enabled, it is noted in the dose tables. In addition, for some organs, such as the uterus, only one dosimeter was used to measure dose. In thes e instances, two dashes are used in place of a number for standard deviation in all dose tables. The head protocol organ doses are present first, and body protocol doses second. Organ doses that result from body scans are broken up into two different sect ions. The first section is devoted entirely to examining differences in primary organ doses with increasing beam width size. Organ doses are presented in 8.4.1 for four different detector configurations, which utilize both helical and ultra helical acquisition modes. Four subjects of varying body ha bitus, were used to determine the effect of varying the scanner detector configuration on organ doses in the primary scan range. The second section displays results for 8 subjects and 5 CT protocols, all at a detector configuration of 0.5 mm x 64. It is w ith the doses presented in this section, that size parameter correlations were analyzed. 8.4.1 Head Protocol Doses Organ doses measured for the head CT protocols described in S ections 6.2.3 and 6.3.1 above are shown in Tables 8 12, 8 13, and 8 14 for the s tandard head protocol, CTA head protocol, and the brain perfusion protocol respectively

PAGE 104

104 For the standard head protocol, the average dose to the brain ranged from 23 mGy to 40 mGy for all detector configurations and all three subjects. The average dos e to the lens from this protocol fell within the 34 57 mGy range for all detector configurations. Finally, the overall average dose to the skin for all detector configurations was between 33 mGy and 58 mGy. These doses are shown illustratively in Figure 8 4 In regards to the CTA head protocol, the average organ doses were found to be within the 70 121 mGy range, 11 56 mGy range, 100 163 mGy range, and 102 164 mGy range for the brain, thyroid, lens, and skin respectively For the brain perfusion proto col, the average organ doses fell within the following ranges: 150 192 mGy, 13 28 mGy, 272 330 mGy, and 272 319 mGy for the brain, thyroid, lens, and skin respectively. Table 8 15 shows the lens dose reduction which can be achieved with the use of a bismuth shield. It must be noted that the lens dose from the brain perfusion protocol with a shield in place for cadaver 6 is only a result of one dose measurement, as opposed to two f or all other doses listed. After that exam was completed, it was discovered that one of the dosimeters had fallen off of the eye of the subject and could not be used to accurately estimate dose. The use of a bismuth shield reduced the lens dose of both s ubjects from 272 mGy and 297 mGy to 206 mGy and 237 mGy. 8.4.2 Body Protocol Organ Doses 8.4.2.1 Organ doses for 4 subjects from helical and ultra helical acquisitions As described in Chapter 6, organ doses for 7 body CT protocols were measured for four di fferent detector configurations, one of which was helical (0.5 mm x 64) and three of which were ultra helical (0.5 mm x 80, 0.5 mm x 100, 0.5 mm x 160). The resulting organ doses from these modes of acquisition are displayed in Tables 8 16 8

PAGE 105

105 22 for the CA P, chest, abdomen, pelvis, three phase liver, pulmonary embolism, and trauma protocols, respectively. Scanning from the thoracic inlet to the lesser trochanter, the CAP protocol irradiates most of the radiosensitive organs for which doses were measured. F or all four subjects, the average organ dose ranged from 8.5 mGy to 33.1 mGy. The maximum organ dose measured was a skin dose of 36.6 mGy, which resulted from one of the largest subjects (cadaver 7) and the 0.5 mm x 160 nominal beam width. Across the boa rd, the lowest average organ doses were measured for the smallest subject (cadaver 4) and the smallest nominal beam width, 0.5 mm x 64. The chest protocol required doses to be measured for the thyroid, breasts, lungs, liver, stomach, and skin. Average o rgan doses for this protocol ranged from 8.2 mGy to 29.5 mGy. The minimum average organ dose of 8.2 mGy was measured for the stomach of the smallest subject (cadaver 4) with the 0.5 mm x 64 detector configuration. The highest dose measured was 30.6 mGy o n the surface of cadaver 5 for the largest detector configuration. The organs examined for the abdomen protocol include the breasts, lungs, liver, stomach, small intestine, and colon. Due to the anatomical positioning of this scan, the lungs are not fully located in the primary beam; therefore, average doses are reported for the entire organ and for just the lower portion of the lungs. For this protocol, the breasts are also subject to these conditions, but were not able to be divided in a similar manner. Thus, the breast dose reported does include all dosimeters that measured breast dose. Average organ doses ranged from 4.7 mGy to 30.0 mGy for the abdomen protocol. The minimum average organ dose of 4.7 mGy was measured for breast of

PAGE 106

106 cadaver 7 and the n ominal beam width of 0.5mm x 64. This is one example of organ variability among subjects. The maximum measured dose for this protocol was 35.5 mGy, which was a skin dose from cadaver 6 and the largest detector configuration. The scan range of the pelvis protocol is the shortest of all protocols examined. Doses were measured for the small intestine, colon, ovaries, uterus, and skin for the pelvis protocol. Average organ doses fell between 5.0 mGy and 28.7 mGy. The minimum average organ dose of 5.0 mGy w as measured for the small intestine of cadaver 7 for the 0.5 mm x 64 detector configuration. The tube for this organ was located just outside the scan range selected by the technologist. The average dose to the small intestine of this subject does increa se with increasing beam widths, representing the effect of over ranging. The maximum measured dose was observed with skin of cadaver 7 for the largest detector configuration. The helical portion of the three phase liver exam is identical in scan range to that of the abdomen protocol. Hence, dose measurements were conducted for the same organs. The average organ doses for the three phase liver protocol ranged from 15.3 mGy to 84.3 mGy, a wide range which encompasses organs in the primary beam and those just outside the selected scan range for some subjects. The maximum dose measured was a 100.5 mGy dose to the skin of cadaver 7 for the 0.5 mm x 100 detector configuration. The pulmonary embolism protocol measured doses for the same organs as the ch est protocol. For this protocol, average organ doses fell within the 12.0 37.7 mGy range. The minimum average organ dose was measured for the breasts of cadaver 4

PAGE 107

107 and the 64 slice scan. The maximum dose measured was 41.6 mGy, a dose to the skin of cadav er 5 with the 0.5 mm x 160 configuration. The maximum amount of dosimeters was used for the trauma protocol, which encompasses the head, cervical spine, and the entire torso. For the head portion of the scan, the minimum average organ dose measured was 29.0 mGy to the brain of cadaver 5 and the 160 slice scan. The maximum measured dose to the head was a lens dose of 81.0 mGy for cadaver 7 and the 64 slice scan. The cervical spine and torso scans resulted in an average organ dose range of 22.4 107.8 mGy The maximum organ dose was a 115.8 mGy dose to the thyroid of cadaver 6 and the 0.5 mmx 160 detector configuration. The minimum average organ dose of 22.4 mGy was a dose measured to the stomach of the smallest subject (cadaver 4) and with the smallest nominal beam width (0.5 mm x 64). In order to more clearly determine the effects of increasing beam width on organ doses, percent differences were calculated for each protocol relative to the 0.5mm x 64 average organ doses. Table 8 23 presents the percen t differences for overall average organ doses for each subject. Tube current modulation was used for all body protocols examined. The tube current modulation algorithm responds differently to various body habitus and beam widths. Figures 8 5, 8 6, and 8 7 display tube current modulation plots for various protocols and detector configurations with all four subjects. 8.4.2.2 Organ doses for 8 subjects from 0.5 mm x 64 acquisitions The direct measurement of organ doses for five common CT protocols has bee n accomplished with 8 cadaveric subjects for the 0.5 mm x 64 detector configuration, the current clinical standard. The organ doses for these 8 subjects are presented in Tables

PAGE 108

108 8 24 8 28 for the CAP, chest, abdomen, pelvis, and three phase liver protocols respectively. The organ doses for the 0.5 mm x 64 detector configuration for subjects 4 7 were presented in 8.4.1 first and are replicated in the tables presented in this section. The organ doses presented in these tables for subjects 1 3 as defined in Table 8 1, are from the dissertation of Thomas Griglock. 79 It must be noted that dose to the uterus was not measured for these subjects, and this is indicated in the tables by the use of dashed lines. The CAP protocol resulted in average organ doses with in the 8.5 41.2 mGy range for all 8 subjects. The maximum organ dose of 41.2 mGy was a dose received by the small intestine of cadaver 1. Initially, it was expected that the maximum organ dose would go to the skin due to its proximity to x ray tube. How ever, upon further evaluation of the tube and dosimeter placement in this subject it was observed that the small intestine tube was placed more inferior than most others, which resulted in closer proximity to pelvic region where tube current is often at a maximum. In addition, skin dosimeters were placed in slightly different positions for the first three subjects, as described in 6.2.1. Only one surface dosimeter was placed over the pelvic region for this subject, which could have resulted in the discrep ancy between the small intestine and skin doses observed for this subject. Average organ doses for the chest protocol fell between 4.7 mGy and 29.7 mGy. The minimum average organ dose was observed for the thyroid of subject 3. Due to anatomical variati ons among subjects, organ doses can vary by a substantial amount for organs at the edge of the scan range. The thyroid of subject 3 was just outside the

PAGE 109

109 selected scan range for this protocol. The maximum dose was a surface dosimeter which measured 36.6 m Gy on subject 1. The abdomen protocol resulted in average organ doses ranging from 1.8 mGy to 30.5 mGy for all 8 subjects. The smallest average organ dose of 1.8 mGy was measured for the lungs of subject 2, which proved to be outside of the scan range due to the location of organ and dosimeters within that subject. The maximum measured dose was 34.2 mGy, to the skin of cadaver 7. The range of average organ doses for the pelvis protocol was between 5.0 mGy and 29.9 mGy. The smallest organ dose was again of the scan range, as selected by the technologist. The highest dose measured for the pelvis protocol was a dose to the skin of 35.8 mGy for cadaver 7. The organ doses measured for the three phase liver protocol were characteristically higher than the doses measured for the other single phase studies examined. The average organ doses ranged from 4.6 mGy to 82.5 mGy. As with the abdomen protocol, the minimum average organ dose was measured for the lungs of subject 2. The maximum dose measured for the three phase liver protocol was 99.7 mGy, from the skin of cadaver 1. 8.4.3 Fetal Dose Estimates The results for each of the three pregnant patient protocols examined and the four cadaveric subjects are shown in Figure 8 8 The average dose to the uterus from a pulm onary embolism CT examination was under 1 mGy for all subects In this case, the uterus is outside of the primary beam. For the CAP and trauma protocols, the fetus is in the primary beam; the fetal dose estima tes ranged from 14.1 mGy to 22.5 mGy for a CAP protocol, and from 25.7 mGy to 40.6 mGy for a trauma protocol.

PAGE 110

110 8.5 S ize S pecific D ose E stimates In order to compare size specific dose estimates with direct dose measurements, the anteri or posterior and lateral dimensions were measured at various locations within the CAP scans for each subject. The effective diameter for each representative AP and lateral dimension was calculated, and the cor responding correction factor was obtained from AAPM 204. The correcti on factor was applied to the exam CTDI Vol which was then compared with the measured organ doses for each cadaver. The first SSDE was calculated based on the dimension measurements at the central slice of the study In order to perform a more detailed co mparison between the SSDEs and organ doses, the second set of SSDEs were calculated based on the central slice of each organ within the CAP scan. Tables 8 29 through 8 36 display the SSDE comparisons with the measured organ doses for all 8 subjects. The SSDE calculated for the skin dose comparison was the same as the central slice SSDE. The central slice SSDE is located in the last row of each table. The average organ dose listed in that row of each table is the overall average of all organ doses listed 8.6 Size Parameter C orrelations Chapter 7 explored the methods utilized to determine the size parameter of interest in this research, the effective diameter. Table 8 37 displays the effective diameters calculated from the measurement of the AP and Lateral d imensions for the central slice of a CAP scan and the central slice of each organ within a CAP scan for all 8 subjects. Correlations were first explored between the average organ doses listed in Table 8 24 and the effective diameters in Table 8 37. A se cond order polynomial was determined through regression methods as the best overall fit to the data. Figure 8 9

PAGE 111

111 illustrates the relationship between the central slice effective diameter and corresponding organ doses. Pearson correlation coefficients ( R 2 ) for Figure 8 9 ranged from 0.595 to 0.924. Figure 8 10 displays the relationship between the central effective diameter of each organ and the corresponding organ doses. R 2 values for Figure 8 10 ranged from 0.578 to 0.893, indicating that overall, using the effective diameter of the central slice results in slightly better correlations than using the central effective diameter of each individual organ. The equations for the correlations observed in Figure 8 9 are displayed in Table 8 38, with the corresp onding R 2 values. In addition to investigating correlations between organ doses and size parameters, scan parameters were also included in the analysis. This was done so that patient specific correlations could be established that not only take into account the size that effect, organ dose conversion factors were calculated by taking the ratio of each average organ dose to the exam CTDI Vol Correlations for these dose conversion factors and the two effective diameters were studied for each organ and are shown in Figures 8 11 and 8 12. By analyzing the overall fit of the data, it was determined that using the effective diameter of the central slice of a scan resu lted in the strongest correlation with the dose conversion coefficient. For that reason, further correlations between organ doses and the central effective diameter of each organ were excluded. The R 2 values from the use of CTDI Vol as the normalization f actor for organ dose ranged from 0.301 to 0.989 for the CAP protocol. Table 8 39 presents the polynomial fit equations correlating organ dose to CTDI Vol conversion coefficients with the central slice effective diameter for the CAP

PAGE 112

112 protocol. While most or gans show a strong correlation between the dose coefficients and effective diameter, a weak correlation is observed for the thyroid, stomach and ovary. The correlation coefficents (R 2 ) and corresponding p values for these three organs do not represent a s tatistically significant correlation. Tables 8 40 through 8 43 present the best fit equations (second order polynomials) for the chest, abdomen, pelvis and three phase liver protocols. For the chest protocol, all organ dose conversion coefficients c orrelated fairly well with effective diameter, except the thyroid, which showed no real correlation at all. The abdomen protocol did not show strong correlations for the lungs and the small intestine. The pelvis protocol had weak correlations for the ova ry and the colon. Finally, the three phase liver protocol showed weak correlations for the lungs and small intestine, the same as the abdomen. With the incorporation of CTDI Vol as a dose normalization factor, there lies a possibility to expand beyond just the protocols for which doses were directly measured. In order to determine if the application of these equations could be broadened to include additional protocols, a trauma protocol was utilized as a test case. Measured organ doses for this protocol w ere compared to the calculated values obtained with the equations from Table 8 39. Since the body portion of a trauma protocol consists of a CAP and AP scan, the CTDI Vol for both of the individual helical scans was extracted from the dose report. Then, t he AP and lateral dimensions were measured at the central slice of both scans, and two values of effective diameter were calculated. Organ doses for each helical scan were calculated separately, by obtaining the organ dose conversion factors and multiplyi Vol The sum of the calculated

PAGE 113

113 organ doses from the CAP and AP scans was compared with the directly measured organ doses. This was done for an overweight and obese patient and the results are presented in Tables 8 44 and 8 45. Anot her possible application of these organ dose equations is the use of only the CAP protocol equations to estimate organ doses for all protocols. If additional subjects are utilized for organ dose measurements, the measurement time would become much more ef ficient if only a CAP protocol had to be studied. For this reason, comparisons between measured organ doses and those calculated from just the CAP equations were explored. This comparison was done for three subjects of varying size and for the chest, abd omen, and pelvis protocol individually. The results are presented in Tables 8 46, 8 47, and 8 48.

PAGE 114

114 T able 8 1. Cadaver body mass index (BMI) classification. Subject Height (in) Weight (lbs) BMI Classification Cadaver 1 67 170.0 26.6 Overweight Cadaver 2 68 143.0 21.7 Normal Cadaver 3 65 100.0 16.6 Underweight Cadaver 4 59 86.0 17.4 Underweight Cadaver 5 65 163.0 27.1 Overweight Cadaver 6 64 255.6 43.9 Extreme Obesity Cadaver 7 62 192.4 35.2 Obese Cadaver 8 62 162.8 29.8 Overweight Table 8 2 Half value layer (HVL) measurements (mm Al) for the AquilionONE Energy/Bowtie Filter 80 kVp 100 kVp 120 kVp 135 kVp Bowtie Filter: S M/L S M/L S M/L S M/L HVL 1 4.13 4.85 4.97 5.89 6.23 7.19 7.04 8.05 HVL 2 5.76 6.76 7.37 8.01 8.87 9.38 9.55 10.23 Homogeneity Coefficient 0.72 0.72 0.67 0.74 0.70 0.77 0.74 0.79 Table 8 3 Beam width and geometric efficiency. Detector Configuration Nominal Width (mm) Measured Width (mm) Penumbra (mm) Geometric Efficiency 0.5 mm x 32 16 25 9 0.64 0.5 mm x 64 32 42.6 10.6 0.75 0.5 mm x 80 40 48.6 8.6 0.82 0.5 mm x 100 50 58.8 8.8 0.85 0.5 mm x 160 80 88 8 0.91 0.5 mm x 320 160 176.4 16.4 0.91

PAGE 115

115 Table 8 4 Scan parameters and over ranging lengths for the head protocol at various detector configurations Scan Parameters 0.5 mm x 32 0.5 mm x 64 0.5 mm x 80 0.5 mm x 100 0.5 mm x 160 kVp 120 120 120 120 120 mA 270 270 270 270 270 Rotation Time (s) 0.75 0.75 0.75 0.75 0.75 Range (mm) 160 160 160 160 160 D FOV 220 S 220 S 220 S 220 S 220 S Eff mAs 309 317 318 233 234 Total Scan Time (s) 13.3 7.6 6.3 4.2 3.2 Focal Spot Small Small Small Small Small Helical Pitch 21 41 51 87 139 CTDI Vol (mGy) 70.8 65.1 64.7 46.3 43.0 DLP (mGy*cm) 1258.7 1231.8 1330.3 1069.8 1164.7 Pitch Factor (p) 0.66 0.64 0.64 0.87 0.87 Nominal Beam Width (n x T) 16 32 40 50 80 TG111: Total Length (mm) 186.20 207.73 214.20 243.60 296.53 OR Length (mm) 26.20 47.73 54.20 83.60 136.53 OR Length (cm) 2.62 4.77 5.42 8.36 13.65 # of OR Rotations 2.50 2.33 2.13 1.92 1.96 DLP/CTDI: Total Range (cm) 17.78 18.92 20.56 23.11 27.09 OR Range (cm) 1.78 2.92 4.56 7.11 11.09 # of OR Rotations 1.69 1.43 1.79 1.63 1.60 Diff b/t 2 methods (cm) 0.84 1.85 0.86 1.25 2.57

PAGE 116

116 Table 8 5 Scan parameters and over ranging lengths for the chest protocol at various detector configurations. Cadaver 4 Cadaver 5 Cadaver 6 Cadaver 7 Scan Parameters 64 80 100 160 64 80 100 160 64 80 100 160 64 80 100 160 kVp 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120 mA R210 R195 R225 R210 R425 R415 R380 R405 R500 R500 R500 R500 R500 R500 R500 R500 Ro tation Time (s) 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Ra nge (mm) 360 360 360 360 360 360 360 360 360 360 360 360 360 360 360 360 Fi lter L L L L LL LL LL LL LL LL LL LL LL LL LL LL Eff mAs 95 123 146 141 169 181 181 252 169 181 180 252 169 181 180 252 To tal Scan Time (s) 4.3 3.9 3.3 2.8 4.3 3.9 3.3 3.1 4.3 3.9 3.3 3.1 4.3 3.9 3.3 3.1 He lical Pitch 95 111 139 159 95 111 139 159 95 111 139 159 95 111 139 159 CT DI Vol (mGy) 5.6 7 7.2 6.9 12.4 12.4 12.1 15.3 15.2 16.2 14.9 20.1 15.8 16.7 15.6 19.8 DL P (mGy*cm) 218.9 278.3 295.9 300.2 486.2 522.4 526.5 720.8 598.6 681.1 650.4 947.4 620.7 673.2 680.3 934.3 Pi tch Factor (p) 1.48 1.39 1.39 0.99 1.48 1.39 1.39 0.99 1.48 1.39 1.39 0.99 1.48 1.39 1.39 0.99 n x T 32 40 50 80 32 40 50 80 32 40 50 80 32 40 50 80 TG 111: To tal Length (mm) 408.5 432.9 458.7 445.2 408.5 432.9 458.7 492.9 408.5 432.9 458.7 492.9 408.5 432.9 458.7 492.9 OR Length (mm) 48.5 72.9 98.7 85.2 48.5 72.9 98.7 132.9 48.5 72.9 98.7 132.9 48.5 72.9 98.7 132.9 OR Length (cm) 4.9 7.3 9.9 8.5 4.9 7.3 9.9 13.3 4.9 7.3 9.9 13.3 4.9 7.3 9.9 13.3 # of OR Rotations 1.0 1.3 1.4 1.1 1.0 1.3 1.4 1.7 1.0 1.3 1.4 1.7 1.0 1.3 1.4 1.7 DL P/CTDI: Total Range (cm) 39.1 39.8 41.1 43.5 39.2 42.1 43.5 47.1 39.4 42.0 43.7 47.1 39.3 40.3 43.6 47.2 OR Range (cm) 3.1 3.8 5.1 7.5 3.2 6.1 7.5 11.1 3.4 6.0 7.7 11.1 3.3 4.3 7.6 11.2 # of OR Rotations 0.7 0.7 0.7 0.9 0.7 1.1 1.1 1.4 0.7 1.1 1.1 1.4 0.7 0.8 1.1 1.4 Di ff b/t 2 methods (cm) 1.8 3.5 4.8 1.0 1.6 1.2 2.4 2.2 1.5 1.2 2.2 2.2 1.6 3.0 2.3 2.1

PAGE 117

117 Table 8 6 Scan parameters and over ranging lengths for the abdomen protocol at various detector configurations. Cadaver 4 Cadaver 5 Cadaver 6 Cadaver 7 Scan Parameters 64 80 100 160 64 80 100 160 64 80 100 160 64 80 100 160 kVp 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120 mA R165 R203 R218 R218 R380 R380 R395 R395 R485 R445 R500 R500 R470 R465 R500 R500 Rotation Time (s) 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Range (mm) 250 250 250 250 300 300 300 300 300 300 300 300 300 300 300 300 Filter L L L L LL LL LL LL LL LL LL LL LL LL LL LL Eff mAs 133 157 167 167 302 308 288 288 302 308 288 288 302 308 288 288 Total Scan Time (s) 5.51 4.28 3.51 2.41 6.6 5.5 4.5 3.1 6.6 5.5 4.5 3.1 6.6 5.5 4.5 3.1 Helical Pitch 53 69 87 139 53 65 87 139 53 65 87 139 53 65 87 139 CTDI Vol (mGy) 7.4 8.4 8.6 8.4 19.6 18.4 21.0 19.7 24.8 24.1 25.6 24.1 25.9 25.1 26.8 25.0 DLP (mGy*cm) 206.5 244.7 257.3 273.5 649.9 639.2 781.6 809.6 819.1 839.2 950.6 991.2 855.4 871.8 997.0 1028.8 Pitch Factor (p) 0.83 0.86 0.87 0.87 0.83 0.81 0.87 0.87 0.83 0.81 0.87 0.87 0.83 0.81 0.87 0.87 n x T 32 40 50 80 32 40 50 80 32 40 50 80 32 40 50 80 TG111: Total Length (mm) 292.0 295.3 305.4 335.0 349.8 357.5 391.5 430.9 349.8 357.5 391.5 430.9 349.8 357.5 391.5 430.9 OR Length (mm) 42.0 45.3 55.4 85.0 49.8 57.5 91.5 130.9 49.8 57.5 91.5 130.9 49.8 57.5 91.5 130.9 OR Length (cm) 4.2 4.5 5.5 8.5 5.0 5.8 9.2 13.1 5.0 5.8 9.2 13.1 5.0 5.8 9.2 13.1 # of OR Rotations 1.6 1.3 1.3 1.2 1.9 1.8 2.1 1.9 1.9 1.8 2.1 1.9 1.9 1.8 2.1 1.9 DLP/CTDI: Total Range (cm) 27.9 29.1 29.9 32.6 33.2 34.7 37.2 41.1 33.0 34.8 37.1 41.1 33.0 34.7 37.2 41.2 OR Range (cm) 2.9 4.1 4.9 7.6 3.2 4.7 7.2 11.1 3.0 4.8 7.1 11.1 3.0 4.7 7.2 11.2 # of OR Rotations 1.1 1.2 1.1 1.1 1.2 1.5 1.7 1.6 1.1 1.5 1.6 1.6 1.1 1.5 1.7 1.6 Diff b/t 2 methods (cm) 1.3 0.4 0.6 0.9 1.8 1.0 1.9 2.0 2.0 0.9 2.0 2.0 2.0 1.0 1.9 1.9

PAGE 118

118 Figure 8 1. Thyroid doses for a standard head protocol at different detector configurations. Figure 8 2. Small intestine and colon doses for a chest protocol at different detector configurations. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 0.5 x 32 0.5 x 64 0.5 x 80 0.5 x 100 0.5 x 160 Average Organ Dose (mGy) Detector Configuration (mm) Standard Head Protocol CAD 5 Thyroid CAD 6 Thyroid CAD 7 Thyroid 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 0.5 x 64 0.5 x 80 0.5 x 100 0.5 x 160 Average Organ Dose (mGy) Detector Configuration (mm) Chest Protocol CAD4 SI CAD5 SI CAD6 SI CAD7 SI CAD4 Colon CAD5 Colon CAD6 Colon CAD7 Colon

PAGE 119

119 Figure 8 3. Upper lung and uterus doses for a n abdomen protocol at various detector configurations. Table 8 7 Calculated f factors as a function of HVLs for each energy and filter combination used in clinical protocols. Energy/ Filter HVL (mm Al) Effective energy (keV) ( en tissue (cm 2 /g) ( en air (cm 2 /g) f factor 80 S 4.13 39.05 0.0806 0.0764 1.06 80 M/L 4.85 41.96 0.0666 0.0630 1.06 100 S 4.97 42.60 0.0647 0.0612 1.06 100 M/L 5.89 46.62 0.0532 0.0502 1.06 120 S 6.23 47.81 0.0499 0.0470 1.06 120 M/L 7.19 51.23 0.0423 0.0397 1.06 135 S 7.04 50.39 0.0432 0.0406 1.06 135 M/L 8.05 55.46 0.0376 0.0352 1.07 Table 8 8 O ptically stimulated luminescent dosimeter (O SLD ) surface correction factors for each energy and filter combination ut ilized for adult head and body computed tomography (C T ) protocols. Surface 120 S 120 M 120 L 135 S 135 M Dosimeter dose (mGy) 26.13 21.70 21.47 30.96 26.95 Ion Chamber air kerma (mGy) 29.34 24.97 25.20 37.00 32.31 Ratio of dosimeter dose to ion chamber air kerma 0.89 0.87 0.85 0.84 0.83 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 0.5 x 64 0.5 x 80 0.5 x 100 0.5 x 160 Average Organ Doses (mGy) Detector Configuration (mm) Abdomen Protocol CAD4 Lung CAD4 Uterus CAD5 Lung CAD5 Uterus CAD6 Lung CAD6 Uterus CAD7 Lung CAD7 Uterus

PAGE 120

120 Table 8 9 OSLD organ correction factors for each energy and filter combination utilized for adult head CT protocols. Head Phantom 120 S 120 M 120 L 135 S 135 M Center 0.91 0.92 0.94 0.92 0.90 Periphery 0.93 0.89 0.87 0.82 0.85 C E,S 0.92 0.91 0.91 0.87 0.87 Table 8 10 OSLD organ correction factors for each energy and filter combination utilized for adult body CT protocols. Body Phantom 120 S 120 M 120 L Center 0.93 0.87 0.92 Middle 0.90 0.91 0.92 Periphery 0.92 0.86 0.84 C E,S 0.92 0.88 0.90 Table 8 11. Validation of surface energy correction factors for body CT protocols. Energy/Filter Measured Dose Corrected Dose Ion Chamber Dose % Difference 120 L 24.47 28.79 30.14 4.48 26.00 30.59 30.14 1.50 26.17 30.79 30.14 2.17 26.89 31.64 30.14 4.98 Average 25.88 30.45 30.14 1.03

PAGE 121

121 Table 8 12 Organ doses for a standard head protocol at various detector configurations. Organ Dose (mGy) Brain Lens Skin Avg Max SD Avg Max SD Avg Max SD Cadaver 5 0.5 x 32 34.7 37.5 2.0 57.1 60.5 4.9 58.2 61.7 3.1 0.5 x 64 32.7 34.3 1.1 54.7 54.7 0.0 53.2 53.8 0.6 0.5 x 80 30.7 34.9 3.8 48.0 52.4 6.2 55.0 57.2 3.2 0.5 x 100 25.1 26.7 1.4 37.1 38.2 1.5 35.7 37.9 1.9 0.5 x 160 23.4 25.3 3.0 36.3 38.5 3.1 40.2 43.9 4.6 0.5 x 320 26.7 30.2 4.0 43.4 44.3 1.2 43.6 44.8 1.0 Cadaver 6 0.5 x 32 34.3 38.1 2.6 56.4 56.5 0.1 54.9 58.7 4.7 0.5 x 64 32.1 34.1 2.0 49.9 55.0 7.3 50.6 51.4 0.7 0.5 x 80 33.3 35.1 1.6 49.5 50.9 2.0 53.5 56.8 3.7 0.5 x 100 27.0 28.9 1.6 34.5 35.1 0.9 35.8 36.0 0.3 0.5 x 160 25.6 28.5 2.1 34.2 35.2 1.4 40.0 44.9 6.2 0.5 x 320 29.3 32.7 4.3 40.9 42.1 1.7 39.4 43.5 4.5 Cadaver 7 0.5 x 32 39.6 41.6 1.4 53.3 55.9 3.7 55.0 60.0 4.4 0.5 x 64 37.1 42.8 3.6 46.3 49.0 3.8 50.0 53.6 3.2 0.5 x 80 39.3 44.8 3.6 39.7 42.1 3.5 51.8 59.2 6.5 0.5 x 100 26.9 32.2 4.9 35.6 37.4 2.6 32.8 36.9 3.6 0.5 x 160 26.6 28.4 1.3 38.9 39.4 0.6 36.0 41.0 4.4 0.5 x 320 37.0 42.9 7.0 45.7 46.5 1.2 37.0 42.9 7.0

PAGE 122

122 Table 8 13 Organ doses for a CT angiography head protocol with a delay scan at various detector configurations. Organ Dose (mGy) Brain Thyroid Lens Skin Subject Avg Max SD Avg Max SD Avg Max SD Avg Max SD Cadaver 5 Shands 101.0 108.8 5.2 11.3 11.8 0.7 161.5 170.0 12.0 162.5 167.6 4.8 0.5 x 64 97.0 100.2 2.9 11.0 11.1 0.2 156.8 158.4 2.3 152.4 154.0 1.7 0.5 x 80 90.1 104.9 12.0 12.4 12.5 0.1 137.5 142.6 7.2 156.3 160.2 6.2 0.5 x 100 73.7 79.3 4.4 10.7 10.9 0.2 110.9 116.0 7.1 103.5 108.7 4.6 0.5 x 160 69.5 76.7 7.7 14.4 14.9 0.6 105.0 105.3 0.4 114.4 122.4 9.0 Cadaver 6 Shands 100.5 109.9 6.6 21.0 21.2 0.3 149.4 156.6 10.2 153.4 167.1 11.9 0.5 x 64 96.2 101.7 4.0 18.8 19.5 1.0 136.3 153.7 24.6 144.7 149.9 8.8 0.5 x 80 98.7 104.2 3.9 22.1 24.2 2.9 140.2 141.5 1.8 149.8 153.6 3.7 0.5 x 100 79.3 82.1 2.1 24.8 25.4 0.8 103.6 104.1 0.7 107.1 110.9 3.5 0.5 x 160 75.1 81.4 5.3 48.8 50.3 2.1 99.9 101.2 1.7 109.9 120.6 11.0 Cadaver 7 Shands 120.6 125.4 3.4 22.4 24.8 3.4 163.0 166.9 5.5 164.4 169.5 4.5 0.5 x 64 115.6 129.8 8.7 22.6 23.7 1.4 149.1 155.9 9.6 154.5 163.8 9.3 0.5 x 80 118.9 136.4 11.7 24.9 24.9 0.0 130.9 135.9 7.1 156.3 176.9 18.1 0.5 x 100 84.0 95.0 10.0 28.4 28.8 0.4 113.5 116.8 4.7 102.1 107.8 7.0 0.5 x 160 82.1 87.0 4.7 55.6 58.0 3.4 114.2 117.4 4.4 109.3 123.7 13.4

PAGE 123

123 Table 8 14 Brain perfusion protocol organ doses. Organ Dose (mGy) Brain Lens Skin Avg Max SD Avg Max SD Avg Max SD Cadaver 5 149.9 173.7 17.6 330.3 331.5 1.7 319.3 340.2 20.4 Cadaver 6 166.7 182.1 10.5 296.7 317.2 29.0 304.0 330.1 24.8 Cadaver 7 191.6 210.0 13.2 272.0 290.0 25.4 271.9 336.0 60.7 Figure 8 4. Average organ doses for various detector configurations with a standard head protocol. Table 8 15 Brain perfusion protocol lens doses with and without a bismuth shield. Lens Dose (mGy) % Reduction Subject Avg Max SD Cadaver 6 296.7 317.2 29.0 20.2 Cadaver 6 w/Shield 236.9 236.9 0.0* Cadaver 7 272.0 290.0 25.4 24.2 Cadaver 7 w/Shield 206.1 209.4 4.6 Only one dosimeter was used to measure dose for the len s of cadaver 6 with the shield. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 0.5 x 32 0.5 x 64 0.5 x 80 0.5 x 100 0.5 x 160 Average Organ Dose (mGy) Detector Configuration CAD5 Brain CAD5 Thyroid CAD5 Lens CAD5 Skin CAD6 Brain CAD6 Thyroid CAD6 Lens CAD6 Skin CAD7 Brain CAD7 Thyroid CAD7 Lens CAD7 Skin CTDI_Vol

PAGE 124

124 Table 8 16. Organ doses from the chest abdomen pelvis (C AP ) protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Skin CAD 4 64 Avg -10.3 11.4 12.2 11.0 14.6 13.5 8.5 14.1 17.0 Max -12.8 14.5 13.3 11.1 14.6 15.0 8.5 14.1 21.6 SD -1.8 2.1 0.8 0.2 -2.2 --2.5 80 Avg -13.1 12.2 14.0 13.6 16.5 15.8 15.6 15.4 18.4 Max -15.8 15.2 14.4 13.8 16.5 15.8 15.6 15.4 22.2 SD -1.4 2.2 0.5 0.3 -0.1 --2.3 100 Avg -12.2 12.1 14.5 13.8 18.1 14.2 11.0 16.5 18.4 Max -15.2 15.0 15.0 14.2 18.1 15.6 11.0 16.5 23.7 SD -2.3 2.0 0.4 0.5 -2.0 --2.9 160 Avg -12.8 11.9 14.1 13.1 18.8 15.9 13.8 15.8 17.8 Max -16.0 15.5 15.2 13.5 18.8 18.0 13.8 15.8 27.5 SD -1.9 1.9 0.7 0.5 -2.9 --4.6 CAD 5 64 Avg 19.7 24.8 18.8 20.3 24.6 28.6 27.5 26.5 22.5 28.1 Max 19.9 27.1 21.2 23.6 25.4 28.6 33.2 26.5 22.5 33.0 SD 0.4 1.8 1.7 3.2 1.0 -8.0 --2.6 80 Avg 17.5 22.6 16.5 19.1 24.4 26.9 28.3 24.5 23.2 25.4 Max 17.5 25.8 20.7 22.0 25.9 26.9 29.3 24.5 23.2 29.3 SD 0.0 3.0 3.4 2.7 2.1 -1.5 --3.0 100 Avg 25.1 24.5 21.2 21.2 25.5 17.6 28.0 23.5 22.8 27.9 Max 26.8 26.0 23.3 24.5 26.6 17.6 28.3 23.5 22.8 32.2 SD 2.4 0.9 1.3 1.9 1.5 -0.4 --2.5 160 Avg 25.2 24.0 19.8 21.3 26.2 22.8 28.2 23.9 23.1 27.4 Max 25.3 25.7 22.5 22.5 27.4 22.8 29.3 23.9 23.1 31.7 SD 0.2 1.2 3.3 0.7 1.8 -1.6 --2.8

PAGE 125

125 Table 8 16. Continued Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Skin 64 Avg 31.0 25.0 21.9 22.8 26.4 29.4 26.4 21.9 22.1 28.4 Max 31.5 27.0 25.0 27.3 28.6 30.9 26.9 24.8 22.1 32.9 SD 0.7 1.7 2.0 2.8 3.1 2.1 0.7 4.1 -3.3 CAD 6 80 Avg 28.6 24.8 20.9 22.3 25.8 26.8 24.7 27.3 23.0 27.4 Max 29.0 29.3 22.4 24.3 27.7 28.1 25.4 27.7 23.0 35.2 SD 0.5 2.2 1.4 1.8 2.7 1.7 1.1 0.5 -3.7 100 Avg 33.1 25.4 22.9 23.4 27.1 27.4 26.1 23.3 23.2 29.4 Max 34.3 27.7 27.5 27.6 28.1 29.5 26.8 27.3 23.2 34.1 SD 1.7 1.1 2.3 4.6 1.4 2.9 1.0 5.6 -3.4 160 Avg 31.6 24.4 22.4 24.1 24.3 29.7 26.0 25.0 21.8 27.7 Max 32.2 29.0 25.3 27.3 28.7 30.7 27.3 26.5 21.8 31.9 SD 0.8 2.2 1.9 2.4 6.1 1.4 1.8 2.1 -3.6 CAD 7 64 Avg 28.4 23.7 18.8 19.3 24.4 22.6 22.9 22.0 19.0 28.8 Max 29.6 28.1 20.9 21.7 25.2 26.7 24.5 24.0 19.0 34.3 SD 1.7 2.2 1.9 1.8 1.1 5.8 2.2 2.8 -4.1 80 Avg 26.0 22.3 18.4 18.8 22.2 21.8 21.2 20.3 18.3 28.7 Max 26.8 24.3 21.2 22.3 24.0 22.3 23.8 20.7 18.3 33.8 SD 1.0 1.5 2.4 3.4 2.6 0.6 3.7 0.7 -3.1 100 Avg 31.3 24.9 21.0 23.3 27.9 22.5 23.8 20.2 21.0 31.1 Max 33.1 28.6 25.9 26.3 28.8 24.2 27.4 20.4 21.0 36.4 SD 2.6 2.2 2.6 2.5 1.2 2.5 5.2 0.2 -3.0 160 Avg 30.4 23.7 19.8 23.2 26.7 23.9 23.3 21.4 19.4 31.4 Max 31.9 28.6 26.7 26.1 29.0 24.8 26.5 23.3 19.4 36.6 SD 2.1 2.3 3.9 2.1 3.2 1.3 4.4 2.6 -3.0

PAGE 126

126 Table 8 17. Organ doses from the chest protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach Skin CAD 4 64 Avg -8.5 8.7 9.9 8.2 9.9 Max -10.3 10.3 10.2 9.6 15.0 SD -1.3 1.1 0.3 1.9 2.0 80 Avg -10.9 10.8 12.1 12.3 11.3 Max -11.9 13.1 12.9 12.3 15.7 SD -1.0 1.3 0.7 0.0 3.7 100 Avg -11.2 11.1 12.3 12.4 12.9 Max -12.3 13.4 13.3 13.0 16.2 SD -1.5 1.2 0.6 0.9 2.2 160 Avg -10.7 10.6 11.2 11.7 12.3 Max -11.6 12.3 11.8 12.2 15.9 SD -0.6 1.1 0.5 0.7 2.0 CAD 5 64 Avg 22.1 15.1 14.9 12.5 13.4 18.9 Max 23.2 18.7 16.3 13.9 14.3 25.2 SD 1.5 2.9 1.0 1.4 1.2 3.6 80 Avg 18.9 16.7 14.8 15.3 14.6 20.3 Max 19.2 20.2 16.7 16.9 14.8 24.5 SD 0.5 2.5 1.8 0.9 0.3 2.9 100 Avg 17.0 17.3 14.4 14.9 16.8 18.9 Max 18.1 20.0 15.5 15.7 17.1 24.3 SD 1.5 1.5 0.9 0.7 0.4 2.5 160 Avg 24.4 22.0 16.1 19.0 20.9 22.9 Max 24.8 23.8 20.6 20.5 21.5 30.6 SD 0.6 1.3 6.5 1.2 0.8 3.5 CAD 6 64 Avg 22.3 16.8 15.0 16.0 12.3 20.7 Max 23.5 20.3 19.3 17.7 12.5 27.2 SD 1.7 3.2 3.0 1.7 0.2 4.4 80 Avg 25.6 15.2 14.7 15.7 12.9 20.4 Max 26.4 20.7 20.0 18.0 13.8 29.5 SD 1.0 3.4 2.3 1.7 1.3 5.1 100 Avg 25.9 16.2 14.5 14.9 13.2 15.6 Max 27.5 19.7 17.5 16.3 13.3 19.5 SD 2.2 2.3 1.6 1.2 0.2 2.5 160 Avg 29.5 21.2 18.8 20.2 16.7 24.0 Max 30.4 24.2 23.4 22.6 20.2 28.2 SD 1.3 3.2 3.0 1.7 4.9 3.3

PAGE 127

127 Table 8 17. Continued Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach Skin 64 Avg 21.2 16.6 14.5 15.1 15.8 20.6 Max 21.5 21.0 18.1 18.4 16.0 28.9 SD 0.3 3.0 2.2 3.0 0.2 4.3 CAD 7 80 Avg 19.6 17.0 14.4 15.1 16.1 21.6 Max 21.7 20.8 17.6 16.1 16.8 28.2 SD 2.9 2.8 1.9 1.1 1.1 5.5 100 Avg 23.2 16.4 13.8 14.6 18.5 19.2 Max 24.5 20.8 18.7 17.8 19.2 24.7 SD 1.9 3.0 3.0 2.9 1.0 2.7 160 Avg 26.3 21.7 18.0 19.3 23.5 24.9 Max 27.6 24.9 24.4 21.6 23.5 28.7 SD 1.9 2.9 3.5 2.4 0.0 2.7

PAGE 128

128 Table 8 18. Organ doses from the abdomen protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Breast Lung Upper Lung Liver Stomach SI Colon Skin CAD 4 64 Avg 9.5 5.6 8.7 9.2 9.7 13.9 8.4 13.4 Max 11.5 10.3 10.3 11.2 9.7 13.9 9.2 16.2 SD 1.3 3.5 1.5 1.8 0.0 -1.1 2.2 80 Avg 11.4 7.3 9.9 12.2 10.6 14.0 10.0 15.5 Max 13.3 13.2 13.2 12.9 11.7 14.0 10.7 18.1 SD 2.0 3.6 3.2 0.5 1.5 -1.1 2.0 100 Avg 11.4 8.5 11.0 12.5 10.4 14.5 9.7 15.6 Max 14.1 13.2 13.2 12.9 11.1 14.5 10.7 18.5 SD 1.8 3.4 2.1 0.4 1.0 -1.4 1.8 160 Avg 10.5 8.2 10.8 12.2 8.9 15.0 10.4 15.2 Max 14.0 13.6 13.6 12.7 10.0 15.0 12.1 18.8 SD 2.6 3.3 2.3 0.5 1.4 -2.4 2.1 CAD 5 64 Avg 20.8 10.3 16.8 19.7 22.1 7.8 21.9 26.0 Max 25.3 17.0 17.0 20.6 22.1 7.8 25.6 32.0 SD 4.1 5.0 0.3 0.7 0.1 -5.2 3.0 80 Avg 20.5 11.2 17.0 19.1 21.3 9.0 23.5 24.1 Max 24.9 17.5 17.5 20.0 23.6 9.0 26.5 29.8 SD 3.0 5.0 0.6 1.2 3.2 -4.2 3.6 100 Avg 22.5 15.1 19.2 21.9 26.1 10.7 27.0 28.7 Max 26.9 20.5 20.5 22.7 27.8 10.7 27.9 34.4 SD 3.6 4.7 1.8 0.5 2.5 -1.3 2.7 160 Avg 23.2 17.1 18.7 20.5 21.5 11.6 26.7 26.8 Max 27.0 19.4 19.2 21.5 26.0 11.6 30.0 31.6 SD 2.7 2.1 0.6 1.4 6.4 -4.6 3.0

PAGE 129

129 Table 8 18. Continued Subject # of Detector Channels Dose (mGy) Breast Lung Upper Lung Liver Stomach SI Colon Skin CAD 6 64 Avg 12.6 8.2 11.9 20.1 22.4 25.6 20.4 27.8 Max 19.6 15.1 15.1 25.2 23.0 25.6 21.9 32.8 SD 6.0 4.3 2.6 3.5 0.8 0.1 2.1 4.2 80 Avg 13.6 8.7 12.3 21.7 22.1 25.1 18.1 28.4 Max 21.6 13.4 13.4 24.6 22.9 25.3 22.7 35.1 SD 6.6 4.0 1.1 2.8 1.1 0.2 6.5 2.9 100 Avg 15.8 10.6 15.1 22.9 25.5 29.9 21.8 29.2 Max 22.8 17.6 17.6 27.7 26.6 30.0 23.5 34.6 SD 5.3 5.0 1.9 3.1 1.5 0.2 2.4 4.2 160 Avg 18.6 13.5 16.0 22.8 21.0 29.6 23.6 29.0 Max 23.6 18.1 18.1 26.1 24.4 30.1 24.8 35.5 SD 3.4 4.6 1.9 2.5 4.8 0.7 1.8 5.6 CAD 7 64 Avg 4.7 6.5 14.5 18.2 19.2 22.7 19.4 28.1 Max 8.9 15.5 15.5 20.7 20.7 25.1 20.6 34.2 SD 2.1 5.3 1.4 1.9 2.1 3.4 1.6 2.6 80 Avg 6.5 7.5 15.5 18.0 19.9 22.6 19.5 27.2 Max 13.3 17.3 17.3 19.7 20.3 26.2 21.7 32.8 SD 3.5 5.6 2.6 1.2 0.6 5.2 3.0 3.1 100 Avg 9.7 9.9 18.1 20.2 26.7 26.3 20.8 30.0 Max 17.1 18.3 18.3 22.4 26.9 28.2 21.6 33.7 SD 5.1 6.4 0.3 1.3 0.2 2.8 1.1 2.8 160 Avg 13.6 11.5 18.6 19.5 24.5 24.7 20.3 29.4 Max 19.7 20.2 20.2 21.4 24.9 25.0 22.8 34.6 SD 5.4 6.8 2.3 1.5 0.7 0.4 3.4 4.5

PAGE 130

130 Ta ble 8 19. Organ doses from the pelvis protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) SI Colon Ovary Uterus Skin CAD 4 64 Avg 9.3 11.2 11.8 12.9 16.6 Max 9.3 12.2 11.8 12.9 20.0 SD -1.4 --1.7 80 Avg 12.7 14.9 14.4 15.4 20.0 Max 12.7 16.6 14.4 15.4 25.7 SD -2.4 --2.8 100 Avg 13.4 14.1 14.9 15.7 20.1 Max 13.4 15.8 14.9 15.7 24.7 SD -2.4 --2.4 160 Avg 13.8 14.0 14.2 11.0 19.8 Max 13.8 14.9 14.2 11.0 23.6 SD -1.3 --1.7 CAD 5 64 Avg 26.4 13.8 22.1 24.4 27.6 Max 26.4 19.3 22.1 24.4 32.5 SD -7.8 --3.4 80 Avg 26.2 17.1 22.0 17.6 26.8 Max 26.2 19.9 22.0 17.6 33.0 SD -4.0 --4.1 100 Avg 24.8 23.2 24.3 22.5 27.0 Max 24.8 24.9 24.3 22.5 31.3 SD -2.4 --3.1 160 Avg 25.8 23.1 16.1 24.0 28.4 Max 25.8 26.6 16.1 24.0 32.0 SD -4.9 --2.6 CAD 6 64 Avg 5.8 14.0 24.3 19.3 26.6 Max 7.5 17.1 24.5 19.3 32.4 SD 2.4 4.3 0.2 -3.6 80 Avg 6.3 15.8 22.3 20.6 27.4 Max 7.2 17.7 24.6 20.6 34.7 SD 1.3 2.7 3.3 -3.6 100 Avg 7.4 18.1 25.7 22.1 28.7 Max 9.5 19.0 26.8 22.1 38.5 SD 3.0 1.3 1.5 -5.0 160 Avg 10.8 19.7 22.1 21.0 28.3 Max 12.1 20.5 25.0 21.0 34.8 SD 1.9 1.2 4.0 -3.9

PAGE 131

1 31 Table 8 19. Continued Subject # of Detector Channels Dose (mGy) SI Colon Ovary Uterus Skin CAD 7 64 Avg 5.0 14.6 23.8 18.0 27.9 Max 5.6 15.2 25.5 18.0 35.7 SD 1.0 0.9 2.3 -5.2 80 Avg 8.7 18.6 18.5 25.6 27.6 Max 10.2 19.1 19.9 25.6 37.4 SD 2.1 0.8 2.0 -5.0 100 Avg 10.4 18.8 25.5 20.3 26.3 Max 12.4 21.0 26.7 20.3 36.0 SD 2.8 3.1 1.7 -4.9 160 Avg 18.4 21.7 22.5 19.0 27.5 Max 20.9 22.5 23.9 19.0 39.1 SD 3.5 1.2 2.1 -6.4

PAGE 132

132 Table 8 20. Organ doses from the three phase liver protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Breast Lung Lower Lung Liver Stomach SI Colon Skin CAD 4 64 Avg 32.1 24.8 35.7 34.2 35.3 39.4 33.8 37.9 Max 34.2 43.4 43.4 35.3 36.3 39.4 33.9 42.6 SD 1.3 12.0 6.9 0.7 1.4 0.2 4.8 80 Avg 30.4 21.3 33.5 33.2 30.7 37.4 29.6 38.2 Max 34.2 35.4 35.4 34.3 33.9 37.4 30.1 44.3 SD 4.0 11.9 2.4 1.2 4.5 -0.7 2.8 100 Avg 30.7 21.3 33.9 32.5 38.1 35.4 31.5 36.5 Max 33.7 34.5 34.5 34.8 40.1 35.4 33.3 44.9 SD 3.2 12.5 0.8 1.5 2.8 -2.5 3.8 160 Avg 28.2 17.6 28.7 29.7 31.3 32.1 25.4 31.6 Max 32.2 29.7 29.7 30.7 32.1 32.1 26.1 35.6 SD 2.2 10.7 1.2 1.0 1.1 -1.0 3.8 CAD 5 64 Avg 59.7 28.2 51.2 49.8 60.9 25.7 68.9 69.5 Max 66.6 52.4 52.4 54.8 66.6 25.7 71.3 90.2 SD 6.6 16.2 1.6 7.0 8.0 -3.4 7.3 80 Avg 54.6 33.7 50.1 51.6 59.0 22.3 68.3 66.4 Max 66.6 53.8 53.8 53.0 62.4 22.3 71.5 74.9 SD 8.9 14.2 5.3 1.0 4.8 -4.6 8.5 100 Avg 62.3 40.7 55.6 59.0 71.0 27.7 78.1 73.4 Max 68.8 57.2 57.2 64.8 73.6 27.7 83.1 86.5 SD 6.0 13.4 2.2 3.6 3.6 -7.0 6.6 160 Avg 59.6 48.4 56.9 55.9 69.1 43.5 70.7 74.8 Max 69.3 59.9 59.9 59.3 70.2 43.5 74.2 87.4 SD 6.8 9.2 4.3 2.4 1.5 -4.9 7.5

PAGE 133

133 Table 8 20. Continued Subject # of Detector Channels Dose (mGy) Breast Lung Lower Lung Liver Stomach SI Colon Skin CAD 6 64 Avg 36.0 23.9 34.3 53.6 60.6 64.1 65.1 82.5 Max 57.0 39.2 39.2 62.1 63.6 77.0 71.4 97.4 SD 17.0 11.5 3.4 5.7 4.2 18.3 8.9 7.8 80 Avg 36.2 24.9 37.4 50.1 49.6 72.3 59.8 72.0 Max 56.1 41.0 41.0 57.8 55.2 73.3 66.7 84.6 SD 15.7 13.8 3.4 4.5 7.9 1.5 9.8 10.2 100 Avg 38.3 26.9 38.7 61.5 68.5 81.0 56.3 82.4 Max 59.5 43.5 43.5 74.6 70.8 83.8 66.7 89.9 SD 16.6 13.0 4.1 8.9 3.3 3.8 14.7 7.0 160 Avg 51.0 34.8 46.5 58.9 69.3 80.3 66.5 78.4 Max 66.7 54.4 54.4 65.5 71.0 81.2 69.1 90.5 SD 13.4 13.2 5.5 4.2 2.4 1.3 3.7 11.3 CAD 7 64 Avg 15.3 19.3 30.5 48.3 56.1 68.3 56.1 78.5 Max 27.2 41.5 41.5 56.3 62.0 75.5 59.7 85.2 SD 6.4 14.6 12.8 5.6 8.3 10.2 5.2 5.4 80 Avg 18.8 21.4 34.0 47.5 58.8 61.2 47.3 74.6 Max 34.3 44.4 44.4 50.3 62.8 63.5 56.7 84.0 SD 8.3 15.4 11.5 1.6 5.7 3.2 13.3 5.5 100 Avg 39.5 33.3 51.1 57.2 76.5 72.8 51.8 80.2 Max 59.7 65.1 65.1 60.6 77.5 74.9 68.7 100.5 SD 16.7 20.7 12.2 3.1 1.4 3.0 23.9 8.5 160 Avg 48.4 36.1 51.2 62.0 69.2 65.4 57.8 84.3 Max 72.8 55.6 55.6 70.1 71.8 65.9 64.9 93.1 SD 15.3 16.5 3.2 7.1 3.7 0.7 10.0 5.7

PAGE 134

134 Table 8 21. Organ doses from the pulmonary embolism protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach Skin CAD 4 64 Avg -12.0 14.0 13.7 15.2 14.7 Max -12.9 15.6 14.6 16.1 18.2 SD -1.3 1.5 1.8 1.4 2.1 80 Avg -14.8 17.9 17.1 18.9 17.3 Max -15.3 19.0 17.9 20.1 21.2 SD -0.5 1.2 0.7 1.7 1.9 100 Avg -14.0 17.0 17.1 17.6 17.7 Max -15.8 22.4 18.8 19.4 24.3 SD -1.7 3.0 1.4 2.6 3.5 160 Avg -16.1 20.2 16.5 18.7 19.7 Max -19.5 29.0 20.8 19.1 23.8 SD -1.7 3.9 4.0 0.6 2.6 CAD 5 64 Avg 18.4 22.8 16.1 21.5 18.6 23.6 Max 18.9 27.8 18.2 22.0 21.0 33.0 SD 0.7 3.3 1.9 0.5 3.5 5.6 80 Avg 16.9 20.9 16.2 20.1 18.5 22.7 Max 17.1 25.8 18.2 21.0 20.2 31.2 SD 0.3 3.7 2.0 0.8 2.3 5.2 100 Avg 23.5 23.9 20.7 23.6 22.7 27.5 Max 24.9 27.8 23.8 25.5 23.7 39.3 SD 2.0 2.5 2.9 1.2 1.3 5.7 160 Avg 21.8 22.8 20.7 22.2 22.5 28.0 Max 22.0 27.8 23.2 23.8 23.0 41.6 SD 0.3 3.9 2.3 1.2 0.7 5.3 CAD 6 64 Avg 12.7 24.5 21.7 20.7 16.7 25.2 Max 13.4 31.1 27.3 25.4 17.1 34.2 SD 0.9 4.0 2.8 2.8 0.7 7.7 80 Avg 31.1 23.5 21.2 18.7 14.3 23.8 Max 31.1 27.3 23.4 20.9 14.4 29.0 SD 0.1 2.1 1.7 1.3 0.1 3.6 100 Avg 32.4 24.2 23.6 21.3 19.9 28.8 Max 33.2 27.0 28.0 24.2 20.1 38.7 SD 1.0 2.8 2.5 3.5 0.4 4.6 160 Avg 30.4 21.9 22.6 22.0 18.3 27.3 Max 32.0 26.1 28.1 24.3 19.4 33.8 SD 2.2 2.8 3.3 1.6 1.5 3.4

PAGE 135

135 Table 8 21. Continued Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach Skin CAD 7 64 Avg 25.8 24.2 18.5 16.8 21.5 27.5 Max 26.1 28.0 20.5 19.0 22.1 33.4 SD 0.4 3.6 1.5 1.4 0.9 3.2 80 Avg 28.1 24.0 18.1 18.5 23.5 28.5 Max 29.4 27.4 21.2 20.7 24.8 33.5 SD 1.8 2.4 2.1 1.5 1.7 3.0 100 Avg 28.4 27.5 20.8 20.2 26.0 28.5 Max 30.1 33.7 25.7 22.4 27.5 37.7 SD 2.4 3.2 3.2 2.2 2.2 4.2 160 Avg 30.6 25.1 20.4 21.2 23.4 26.6 Max 31.1 31.4 23.8 24.4 25.5 32.8 SD 0.7 3.1 2.9 2.7 3.0 3.0

PAGE 136

136 Table 8 22. Organ doses from the trauma protocol for various detector configurations. Subject # of Detector Channels Dose (mGy) Brain Lens Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Skin CAD 4 64 Avg ---22.6 22.6 22.8 22.4 27.2 24.1 23.5 25.7 26.8 Max ---25.9 31.4 24.9 22.4 27.2 24.5 23.5 25.7 35.3 SD ---2.4 4.7 2.3 0.1 -0.6 --5.3 80 Avg ---24.0 24.2 23.1 24.0 32.8 28.5 27.9 29.9 30.2 Max ---26.2 32.4 25.4 25.3 32.8 28.9 27.9 29.9 38.1 SD ---1.8 6.0 2.1 1.8 -0.5 --4.8 100 Avg ---25.4 24.7 24.6 25.5 32.2 31.1 26.9 27.5 33.0 Max ---29.6 32.1 27.4 26.2 32.2 31.2 26.9 27.5 40.3 SD ---2.7 4.3 2.8 0.9 -0.1 --5.4 160 Avg ---25.1 29.5 24.2 26.5 30.8 27.6 27.9 29.1 32.1 Max ---30.3 39.7 29.7 27.1 30.8 28.7 27.9 29.1 46.1 SD ---3.0 5.9 3.7 0.8 -1.6 --5.6 CAD 5 64 Avg 37.2 54.8 29.6 38.7 29.4 35.4 40.5 47.8 49.6 40.8 40.6 51.4 Max 42.6 62.6 33.1 48.5 32.3 36.2 43.4 47.8 56.1 40.8 40.6 66.1 SD 3.4 10.9 5.0 6.6 2.3 1.0 4.1 -9.2 --7.3 80 Avg 37.9 67.2 55.0 38.9 28.9 34.3 40.0 45.7 40.8 37.5 40.1 47.9 Max 42.0 71.8 57.2 46.0 31.2 35.9 42.5 45.7 51.3 37.5 40.1 61.4 SD 2.9 6.6 3.1 5.4 1.8 1.0 3.5 -14.9 --7.8 100 Avg 29.4 44.9 58.4 48.8 37.6 38.8 49.2 54.4 54.8 32.2 46.3 54.7 Max 32.7 46.8 61.0 52.8 40.6 44.9 53.7 54.4 60.5 32.2 46.3 68.1 SD 2.1 2.6 3.7 3.0 2.8 4.7 6.5 -8.0 --9.0 160 Avg 29.0 50.9 63.7 49.6 42.3 46.5 50.4 51.6 51.5 41.9 41.4 53.7 Max 34.1 53.8 65.0 59.4 48.9 52.3 52.5 51.6 57.2 41.9 41.4 67.8 SD 5.4 4.0 1.7 4.1 5.5 3.4 3.0 -8.1 --7.9

PAGE 137

137 Table 8 22. Continued Subject # of Detector Channels Dose (mGy) Brain Lens Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Skin CAD 6 64 Avg 38.1 59.8 81.4 34.5 31.9 42.1 43.0 50.1 43.5 40.4 37.2 51.7 Max 39.9 60.6 88.2 42.9 36.4 52.1 43.7 50.5 47.4 41.3 37.2 66.4 SD 1.4 1.0 9.7 4.9 3.4 7.8 1.0 0.6 5.5 1.3 -8.7 80 Avg 38.6 66.7 74.4 36.9 33.5 40.2 42.4 49.0 42.0 42.7 37.7 54.5 Max 43.8 69.3 76.1 46.1 37.9 44.0 45.5 50.2 43.9 46.5 37.7 74.0 SD 3.0 3.7 2.3 4.2 2.2 2.8 4.3 1.6 2.7 5.3 -8.1 100 Avg 30.9 41.2 101.4 33.1 33.7 32.3 35.9 42.5 34.5 35.1 29.5 41.4 Max 32.1 45.2 108.2 37.1 41.2 34.9 37.2 42.8 35.2 35.5 29.5 49.4 SD 1.4 5.7 9.6 2.1 3.5 2.4 1.9 0.4 1.0 0.6 -4.5 160 Avg 32.7 49.5 107.8 37.2 41.9 43.2 44.5 52.6 44.7 44.4 39.4 49.1 Max 34.4 50.7 115.8 45.6 49.8 48.3 45.7 53.1 49.2 45.0 39.4 62.4 SD 1.8 1.8 11.4 4.3 6.4 5.0 1.7 0.8 6.3 0.9 -6.8 CAD 7 64 Avg 39.7 79.4 72.9 47.0 38.3 38.8 50.6 44.5 40.3 40.8 34.8 52.1 Max 45.2 81.0 74.7 65.3 46.9 41.9 53.6 44.5 44.3 41.4 34.8 60.3 SD 5.4 2.3 2.5 10.5 5.8 2.7 4.2 0.1 5.6 0.9 -5.1 80 Avg 40.9 54.6 78.4 54.4 43.4 38.6 54.6 44.3 42.1 40.0 39.3 54.2 Max 48.8 65.5 85.5 67.2 47.8 42.5 57.0 49.7 45.4 43.6 39.3 65.1 SD 6.2 15.4 10.1 9.1 3.6 2.7 3.3 7.6 4.6 5.2 -6.5 100 Avg 31.6 49.7 92.2 59.2 45.2 43.3 59.1 56.8 44.4 44.0 40.2 52.8 Max 38.9 54.5 97.6 74.1 58.0 47.2 63.6 61.4 50.3 45.2 40.2 70.4 SD 5.1 6.8 7.6 6.7 6.2 3.4 6.2 6.6 8.3 1.8 -8.3 160 Avg 31.6 48.6 104.9 60.9 48.7 43.5 58.7 48.7 39.5 39.9 41.4 52.5 Max 36.0 48.7 106.8 67.4 56.0 44.8 60.2 51.6 42.3 40.6 41.4 74.9 SD 3.3 0.2 2.7 5.9 4.6 1.4 2.2 4.1 4.0 0.9 -9.7

PAGE 138

138 Table 8 23. Percent difference in dose for various detector configurations relative to 0.5 mm x 64 average doses. % Dose Difference Relative to 0.5 x 64 Average Organ Doses Protocol Detector Configuration CAD 4 CAD 5 CAD 6 CAD 7 Average for all subjects CAP 0.5 x 80 22.6 6.8 6.5 4.5 9.8 0.5 x 100 17.0 10.3 4.1 8.0 9.7 0.5 x 160 19.0 8.5 4.0 5.9 9.6 Chest 0.5 x 80 27.2 10.7 5.8 2.8 10.9 0.5 x 100 32.5 14.2 10.4 7.1 15.3 0.5 x 160 25.2 32.0 26.9 29.2 28.5 Abdomen 0.5 x 80 17.4 6.2 4.6 2.6 9.3 0.5 x 100 22.2 18.9 13.8 17.7 23.4 0.5 x 160 20.8 15.2 14.9 13.7 27.7 Pelvis 0.5 x 80 26.7 11.2 7.9 33.5 19.8 0.5 x 100 28.0 18.8 16.7 32.8 24.1 0.5 x 160 25.8 20.4 29.9 66.6 35.7 3PL 0.5 x 80 7.4 4.6 11.2 8.1 8.4 0.5 x 100 7.1 11.8 13.4 23.2 20.2 0.5 x 160 18.1 19.4 15.3 22.4 29.7 PE 0.5 x 80 23.8 4.7 29.9 5.9 15.7 0.5 x 100 19.9 18.2 33.5 13.5 21.3 0.5 x 160 31.4 15.0 29.7 11.9 21.6 Trauma 0.5 x 80 12.2 12.9 4.8 8.6 9.4 0.5 x 100 15.1 24.0 17.8 18.3 19.0 0.5 x 160 16.2 24.5 11.5 18.5 17.8 Figure 8 5. Tube current modulation plots of various detector configurat ions for the chest protocol or pulmonary embolism protocol for cadaver 5.

PAGE 139

139 Figure 8 6. Tube current modulation plots of various detector configurations for the abdomen protocol or three phase liver protocol for cadaver 6. Figure 8 7. Tube current modulation plots of the 0.5 mm x 64 and 0.5 mm x 160 detector configurations for the chest abdomen pelvis ( CAP ) protocol for a small (cadaver 4) and large subject (cadaver 7).

PAGE 140

140 Table 8 24. Organ doses from a CAP protocol for 8 subjects at a de tector configuration of 0.5 mm x 64. Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Skin CAD 1 64 Avg 28.2 25.2 20.5 28.7 28.4 41.2 31.4 23.3 -28.4 Max 28.2 26.5 21.7 31.5 28.4 41.2 34.7 23.3 -33.5 SD -1.6 1.0 1.9 --4.7 --3.4 CAD 2 64 Avg 13.9 17.4 11.8 17.6 14.4 17.5 15.8 14.7 -18.2 Max 13.9 18.5 12.6 19.0 14.4 17.5 16.5 14.7 -26.4 SD -1.0 0.6 1.1 --1.0 --3.7 CAD 3 64 Avg 10.0 23.1 15.5 23.8 29.3 27.2 22.1 15.7 -30.3 Max 10.0 25.1 22.3 30.1 29.3 27.2 24.8 15.7 -35.5 SD -1.8 5.2 4.6 --3.8 --3.9 CAD 4 64 Avg -10.3 11.4 12.2 11.0 14.6 13.5 8.5 14.1 17.0 Max -12.8 14.5 13.3 11.1 14.6 15.0 8.5 14.1 21.6 SD -1.8 2.1 0.8 0.2 -2.2 --2.5 CAD 5 64 Avg 19.7 24.8 18.8 20.3 24.6 28.6 27.5 26.5 22.5 28.1 Max 19.9 27.1 21.2 23.6 25.4 28.6 33.2 26.5 22.5 33.0 SD 0.4 1.8 1.7 3.2 1.0 -8.0 --2.6 CAD 6 64 Avg 31.0 25.0 21.9 22.8 26.4 29.4 26.4 21.9 22.1 28.4 Max 31.5 27.0 25.0 27.3 28.6 30.9 26.9 24.8 22.1 32.9 SD 0.7 1.7 2.0 2.8 3.1 2.1 0.7 4.1 -3.3 CAD 7 64 Avg 28.4 23.7 18.8 19.3 24.4 22.6 22.9 22.0 19.0 28.8 Max 29.6 28.1 20.9 21.7 25.2 26.7 24.5 24.0 19.0 34.3 SD 1.7 2.2 1.9 1.8 1.1 5.8 2.2 2.8 -4.1 CAD 8 64 Avg 22.0 20.0 17.9 18.8 19.4 23.8 23.3 21.6 22.0 23.9 Max 22.1 23.4 21.7 20.5 19.5 24.2 25.6 26.9 22.0 28.7 SD 0.2 2.5 2.2 1.8 0.2 0.6 3.3 7.5 -3.5

PAGE 141

141 Table 8 25. Organ doses from a chest protocol for 8 subjects at a detector configuration of 0.5 mm x 64. Subject # of Detector Channels Dose (mGy) Thyroid Breast Lung Liver Stomach Skin CAD 1 64 Avg 21.2 26.9 22.1 23.8 24.3 29.7 Max 21.2 28.2 25.1 26.7 24.3 36.6 SD -1.6 2.2 2.7 -4.0 CAD 2 64 Avg 8.3 14.7 10.2 11.7 10.3 13.6 Max 8.3 15.3 10.9 13.7 10.3 17.4 SD -0.7 0.9 2.1 -3.2 CAD 3 64 Avg 4.7 17.3 13.9 15.3 15.0 18.7 Max 4.7 20.0 16.0 19.3 15.0 22.5 SD -3.4 2.0 3.7 -3.0 CAD 4 64 Avg -8.5 8.7 9.9 8.2 9.9 Max -10.3 10.3 10.2 9.6 15.0 SD -1.3 1.1 0.3 1.9 2.0 CAD 5 64 Avg 22.1 15.1 14.9 12.5 13.4 18.9 Max 23.2 18.7 16.3 13.9 14.3 25.2 SD 1.5 2.9 1.0 1.4 1.2 3.6 CAD 6 64 Avg 22.3 16.8 15.0 16.0 12.3 20.7 Max 23.5 20.3 19.3 17.7 12.5 27.2 SD 1.7 3.2 3.0 1.7 0.2 4.4 CAD 7 64 Avg 21.2 16.6 14.5 15.1 15.8 20.6 Max 21.5 21.0 18.1 18.4 16.0 28.9 SD 0.3 3.0 2.2 3.0 0.2 4.3 CAD 8 64 Avg 18.8 17.8 13.0 15.5 11.1 19.6 Max 18.8 20.9 17.3 19.5 11.7 25.2 SD 0.0 2.2 3.2 4.0 0.8 3.1

PAGE 142

142 Table 8 26. Organ doses from an abdomen protocol for 8 subjects at a detector configuration of 0.5 mm x 64. Subject # of Detector Channels Dose (mGy) Breast Lung Lower Lung Liver Stomach SI Colon Skin CAD 1 64 Avg 17.5 13.4 20.7 24.9 30.0 30.5 28.8 30.1 Max 23.4 22.0 22.0 27.7 30.0 30.5 29.5 33.7 SD 5.3 8.5 1.8 2.7 --1.0 3.1 CAD 2 64 Avg 13.2 1.8 3.1 12.2 12.0 4.6 13.2 15.4 Max 15.1 3.2 3.2 13.2 12.0 4.6 13.7 19.5 SD 1.8 1.4 0.3 1.6 --0.8 2.8 CAD 3 64 Avg 19.4 10.9 18.6 24.3 26.9 18.3 27.5 24.4 Max 22.6 19.3 19.3 26.9 26.9 18.3 29.4 30.9 SD 3.0 9.0 0.9 1.7 --2.7 5.8 CAD 4 64 Avg 9.5 5.6 8.7 9.2 9.7 13.9 8.4 13.4 Max 11.5 10.3 10.3 11.2 9.7 13.9 9.2 16.2 SD 1.3 3.5 1.5 1.8 0.0 -1.1 2.2 CAD 5 64 Avg 20.8 10.3 16.8 19.7 22.1 7.8 21.9 26.0 Max 25.3 17.0 17.0 20.6 22.1 7.8 25.6 32.0 SD 4.1 5.0 0.3 0.7 0.1 -5.2 3.0 CAD 6 64 Avg 12.6 8.2 11.9 20.1 22.4 25.6 20.4 27.8 Max 19.6 15.1 15.1 25.2 23.0 25.6 21.9 32.8 SD 6.0 4.3 2.6 3.5 0.8 0.1 2.1 4.2 CAD 7 64 Avg 4.7 6.5 14.5 18.2 19.2 22.7 19.4 28.1 Max 8.9 15.5 15.5 20.7 20.7 25.1 20.6 34.2 SD 2.1 5.3 1.4 1.9 2.1 3.4 1.6 2.6 CAD 8 64 Avg 15.5 9.9 16.1 17.5 16.9 5.7 15.7 20.9 Max 20.2 17.2 17.2 20.4 17.7 5.8 19.0 26.0 SD 5.3 6.7 0.9 2.5 1.1 0.2 4.6 2.5

PAGE 143

143 Table 8 27. Organ doses from a pelvis protocol for 8 subjects at a detector configuration of 0.5 mm x 64. Subject # of Detector Channels Dose (mGy) SI Colon Ovary Uterus Skin CAD 1 64 Avg 22.2 23.2 25.4 -29.9 Max 22.2 23.2 25.4 -35.3 SD -0.0 --6.2 CAD 2 64 Avg 13.0 10.0 12.1 -17.8 Max 13.0 13.2 12.1 -24.2 SD -4.5 --5.0 CAD 3 64 Avg 24.9 10.5 19.8 -29.2 Max 24.9 10.7 19.8 -31.9 SD -0.3 --2.1 CAD 4 64 Avg 9.3 11.2 11.8 12.9 16.6 Max 9.3 12.2 11.8 12.9 20.0 SD -1.4 --1.7 CAD 5 64 Avg 26.4 13.8 22.1 24.4 27.6 Max 26.4 19.3 22.1 24.4 32.5 SD -7.8 --3.4 CAD 6 64 Avg 5.8 14.0 24.3 19.3 26.6 Max 7.5 17.1 24.5 19.3 32.4 SD 2.4 4.3 0.2 -3.6 CAD 7 64 Avg 5.0 14.6 23.8 18.0 27.9 Max 5.6 15.2 25.5 18.0 35.7 SD 1.0 0.9 2.3 -5.2 CAD 8 64 Avg 23.5 16.4 22.5 19.2 27.0 Max 25.1 17.1 24.0 19.2 33.1 SD 2.3 1.0 2.1 -3.4

PAGE 144

144 Table 8 28. Organ doses from a three phase liver protocol for 8 subjects at a detector configuration of 0.5 mm x 64. Subject # of Detector Channels Dose (mGy) Breast Lung Lower Lung Liver Stomach SI Colon Skin CAD 1 64 Avg 47.4 40.3 63.2 75.8 74.9 57.4 64.6 56.7 Max 58.2 67.4 67.4 84.1 74.9 57.4 71.6 99.7 SD 13.7 26.7 5.9 6.2 --9.9 27.8 CAD 2 64 Avg 29.7 4.6 7.6 37.5 37.1 11.8 36.3 34.5 Max 34.8 7.7 7.7 42.4 37.1 11.8 36.8 42.2 SD 6.6 3.5 0.1 3.8 --0.6 6.3 CAD 3 64 Avg 34.1 31.0 53.5 54.8 66.7 22.0 67.8 67.8 Max 61.6 55.8 55.8 65.1 66.7 22.0 70.3 85.9 SD 23.8 26.0 3.4 8.1 --3.6 11.2 CAD 4 64 Avg 32.1 24.8 35.7 34.2 35.3 39.4 33.8 37.9 Max 34.2 43.4 43.4 35.3 36.3 39.4 33.9 42.6 SD 1.3 12.0 6.9 0.7 1.4 -0.2 4.8 CAD 5 64 Avg 59.7 28.2 51.2 49.8 60.9 25.7 68.9 69.5 Max 66.6 52.4 52.4 54.8 66.6 25.7 71.3 90.2 SD 6.6 16.2 1.6 7.0 8.0 -3.4 7.3 CAD 6 64 Avg 36.0 23.9 34.3 53.6 60.6 64.1 65.1 82.5 Max 57.0 39.2 39.2 62.1 63.6 77.0 71.4 97.4 SD 17.0 11.5 3.4 5.7 4.2 18.3 8.9 7.8 CAD 7 64 Avg 15.3 19.3 30.5 48.3 56.1 68.3 56.1 78.5 Max 27.2 41.5 41.5 56.3 62.0 75.5 59.7 85.2 SD 6.4 14.6 12.8 5.6 8.3 10.2 5.2 5.4 CAD 8 64 Avg 18.6 17.2 30.7 42.4 42.2 13.0 36.2 52.4 Max 40.7 34.6 34.6 47.5 43.4 14.3 42.4 58.1 SD 13.3 14.7 3.9 5.0 1.7 1.7 8.7 4.4

PAGE 145

145 Figure 8 8 Fetal dose estimates for a pulmonary embolism, chest abdomen pelvis, and trauma protocol for four cadaveric subjects. Table 8 29. Size specific dose estimate comparison with measured organ doses for cadaver 1. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 157.2 411.3 25.4 39 28.2 38.5 Breast 201.8 364.4 27.1 36 25.2 42.6 Lung 197.1 362.8 26.7 37 20.5 80.4 Liver 243.2 361.3 29.6 33 28.7 15.2 Stomach 243.2 361.3 29.6 33 28.4 16.0 SI 238.5 401.2 30.9 32 41.2 22.3 Colon 247.9 382.4 30.8 32 31.4 1.9 Ovary 216.6 404.3 29.6 33 23.3 41.4 Skin 241.6 366.8 29.8 33 28.4 16.2 Central 241.6 366.8 29.8 33 28.4 16.3 0 5 10 15 20 25 30 35 40 45 PE CAP TRAUMA Uterus Dose (mGy) Protocol CAD 4 CAD 5 CAD 6 CAD 7

PAGE 146

146 Table 8 30. Size specific dose estimate comparison with measured organ doses for cadaver 2. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 113.7 313.5 18.9 18 13.9 29.2 Breast 191.7 304.7 24.2 15 17.4 15.0 Lung 196.8 296 24.1 15 11.8 25.4 Liver 193.9 293.8 23.9 15 17.6 15.2 Stomach 201.2 294.5 24.3 15 14.4 2.2 SI 187.4 301.8 23.8 15 17.5 14.5 Colon 187.4 301.8 23.8 15 15.8 5.3 Ovary 180.1 325.1 24.2 15 14.7 0.2 Skin 193.9 295.3 23.9 15 18.2 18.3 Central 193.9 295.3 23.9 15 15.7 5.2 Table 8 31. Size specific dose estimate comparison with measured organ doses for cadaver 3. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 104.7 318.7 18.3 29 10.0 189.4 Breast 219.5 358.5 28.1 20 23.1 13.3 Lung 197.6 339 25.9 22 15.5 41.6 Liver 226.5 330.4 27.4 21 23.8 11.8 Stomach 229.6 324.1 27.3 21 29.3 28.2 SI 210.9 356 27.4 21 27.2 22.9 Colon 209.3 346 26.9 21 22.1 5.1 Ovary 198.4 342.9 26.1 22 15.7 39.9 Skin 213.2 331.1 26.6 21 30.3 30.7 Central 213.2 331.1 26.6 21 21.9 4.1

PAGE 147

147 Table 8 32. Size specific dose estimate comparison with measured organ doses for cadaver 4. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Breast 173.4 295.2 22.6 12 10.3 16.3 Lung 167.9 325.7 23.4 12 11.4 5.7 Liver 173.4 268.7 21.6 13 12.2 6.2 Stomach 181.2 285.1 22.7 12 11.0 9.3 SI 182.8 303.0 23.5 12 14.6 17.8 Colon 182.0 303.8 23.5 12 13.5 11.0 Ovary 178.1 353.0 25.1 11 8.5 29.9 Uterus 178.1 353.0 25.1 11 14.1 22.2 Skin 184.3 292.1 23.2 12 17.0 29.4 Central 184.3 292.1 23.2 12 11.9 0.4 Table 8 33. Size specific dose estimate comparison with measured organ doses for cadaver 5. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 154.6 374.9 24.1 28 19.7 42.5 Breast 206.2 390.5 28.4 24 24.8 3.2 Lung 196.8 356.1 26.5 26 18.8 38.1 Liver 212.4 390.5 28.8 24 20.3 18.1 Stomach 218.7 386.6 29.1 23 24.6 6.6 SI 236.7 383.5 30.1 23 28.6 19.6 Colon 228.8 378.8 29.4 23 27.5 16.5 Ovary 222.6 395.2 29.7 23 26.5 13.3 Uterus 215.6 397.5 29.3 23 22.5 2.1 Skin 229.6 336.6 27.8 24 28.1 14.6 Central 229.6 336.6 27.8 24 24.2 0.6

PAGE 148

148 Table 8 34. Size specific dose estimate comparison with measured organ doses for cadaver 6. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 192.3 497.9 30.9 30 31.0 3.2 Breast 227.4 446.2 31.9 29 25.0 16.1 Lung 219.6 447.1 31.3 29 21.9 32.3 Liver 267.4 382.6 32.0 29 22.8 27.2 Stomach 270.4 387.5 32.4 28 26.4 6.0 SI 279.2 435.3 34.9 26 29.4 11.6 Colon 284.0 442.1 35.4 25 26.4 5.3 Ovary 266.4 494.8 36.3 25 21.9 14.0 Uterus 260.6 497.8 36.0 25 22.1 13.1 Skin 272.3 393.3 32.7 28 28.4 1.3 Central 272.3 393.3 32.7 28 25.5 9.7 Table 8 35. Size specific dose estimate comparison with measured organ doses for cadaver 7. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 152.3 481.2 27.1 35 28.4 23.1 Breast 223.5 434.3 31.2 30 23.7 26.7 Lung 232.3 435.3 31.8 30 18.8 59.8 Liver 254.7 423.6 32.8 28 19.3 44.8 Stomach 249.9 397.2 31.5 30 24.4 23.1 SI 253.8 455.8 34.0 27 22.6 19.5 Colon 253.8 455.8 34.0 27 22.9 17.9 Ovary 244.0 490.0 34.6 26 22.0 18.1 Uterus 247.9 490.0 34.9 26 19.0 36.8 Skin 252.8 427.5 32.9 28 28.8 2.8 Central 252.8 427.5 32.9 28 23.0 21.8

PAGE 149

149 Table 8 36. Size specific dose estimate comparison with measured organ doses for cadaver 8. Organ AP (mm) LAT (mm) ED (cm) SSDE (mGy) Avg Organ Dose (mGy) % Difference Thyroid 164.7 417.0 26.2 26 22.0 18.4 Breast 204.2 358.6 27.1 25 20.0 25.2 Lung 196.5 366.4 26.8 25 17.9 39.3 Liver 214.5 353.5 27.5 24 18.8 27.9 Stomach 216.2 360.4 27.9 24 19.4 23.7 SI 211.9 419.6 29.8 23 23.8 3.3 Colon 211.9 419.6 29.8 23 23.3 1.1 Ovary 199.9 429.0 29.3 23 21.6 6.7 Uterus 199.1 431.6 29.3 23 22.0 4.5 Skin 215.4 366.4 28.1 24 23.9 0.5 Central 215.4 366.4 28.1 24 21.3 12.9 Table 8 37. Effective diameters for all subjects from a CAP protocol. Organ Effective Diameters by Subject (cm) CAD 1 CAD 2 CAD 3 CAD 4 CAD 5 CAD 6 CAD 7 CAD 8 Thyroid 28.16 13.93 10.02 -19.65 30.99 28.43 21.97 Breast 25.25 17.36 23.06 10.32 24.81 24.98 23.68 19.97 Lung 20.51 11.78 15.54 11.35 18.83 21.91 18.77 17.94 Liver 28.65 17.59 23.81 12.25 20.33 22.81 19.34 18.76 Stomach 28.44 14.36 29.26 10.98 24.63 26.42 24.37 19.39 SI 41.18 17.51 27.23 14.60 28.59 29.42 22.60 23.79 Colon 31.40 15.80 22.12 13.48 27.54 26.39 22.90 23.26 Ovary 23.33 14.72 15.72 8.47 26.51 21.92 22.02 21.56 Uterus ---14.14 22.52 22.11 19.00 22.02 Skin 28.41 18.21 30.31 17.00 28.12 28.37 28.81 23.89 Center 29.77 23.93 26.57 23.20 27.80 32.73 32.87 28.09

PAGE 150

150 Figure 8 9. Organ dose vs. effective diameter of the central slice for a CAP protocol. Figure 8 10. Organ dose vs. central effective diameter of each organ for a CAP protocol. 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 15.00 20.00 25.00 30.00 35.00 40.00 Organ Dose (mGy) Effective Diameter (cm) Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Poly. (Thyroid) Poly. (Breast) Poly. (Lung) Poly. (Liver) Poly. (Stomach) Poly. (SI) Poly. (Colon) Poly. (Ovary) Poly. (Uterus) 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 20 22 24 26 28 30 32 34 Organ Dose (mGy) Effective Diameter (cm) Breast Lung Liver Stomach SI Colon Ovary Uterus Skin Thyroid Poly. (Breast) Poly. (Lung) Poly. (Liver) Poly. (Stomach) Poly. (SI) Poly. (Colon) Poly. (Ovary) Poly. (Uterus) Poly. (Skin) Poly. (Thyroid)

PAGE 151

151 Table 8 38. Polynomial fit equations for correlations between the effective diameter of the central slice and organ doses for a CAP protocol. Organ Polynomial Fit Equation R 2 Thyroid y = 0.0119x 2 + 1.464x 30.346 0.782 Breast y = 0.2291x 2 + 14.034x 189.74 0.837 Lung y = 0.1171x 2 + 7.576x 102 0.924 Liver y = 0.2422x 2 + 14.385x 189.83 0.595 Stomach y = 0.3067x 2 + 18.576x 254.13 0.730 SI y = 0.438x 2 + 25.949x 352.75 0.647 Colon y = 0.3217x 2 + 19.305x 261.98 0.854 Ovary y = 0.283x 2 + 17.141x 235.71 0.811 Uterus y = 0.2184x 2 + 12.897x 167.53 0.911 Skin y = 0.2185x 2 + 13.396x 176.09 0.784 Figure 8 11. Organ dose to computed tomography dose index (C TDI Vol ) conversion coefficients vs. effective diameter of the central slice of a CAP protocol. 0.00 0.50 1.00 1.50 2.00 2.50 20 25 30 35 Organ Dose Conversion Coefficent Effective Diameter (cm) Breast Lung Liver Stomach SI Colon Ovary Uterus Skin Thyroid Poly. (Breast) Linear (Lung) Linear (Liver) Poly. (Stomach) Poly. (SI) Poly. (Colon) Poly. (Ovary) Poly. (Uterus) Poly. (Skin)

PAGE 152

152 Figure 8 12. Organ dose to CTDI Vol conversion coefficients vs. the central effective diameter of each organ for a CAP protocol. Table 8 39. Polynomial fit equatio ns for correlations between the effective diameter of the central slice and organ dose to computed tomography dose index ( CTDI Vol ) conversion coefficients for a CAP protocol. Organ Polynomial Fit Equation R 2 Standard Error P value Thyroid y = 0.0139x 2 0.8029x + 12.578 0.301 0.245 0.489 Breast y = 0.0006x 2 0.1023x + 3.6733 0.686 0.203 0.055 Lung y = 0.0072x 2 0.4665x + 8.3695 0.902 0.091 0.003 Liver y = 0.0041x 2 0.3256x + 7.1282 0.861 0.167 0.007 Stomach y = 0.0045x 2 + 0.1936x 0.5255 0.502 0.267 0.175 SI y = 0.0012x 2 0.0279x + 3.1905 0.889 0.164 0.009 Colon y = 3E 05x 2 0.0779x + 3.5609 0.941 0.085 0.001 Ovary y = 0.0044x 2 + 0.2028x 1.0242 0.491 0.216 0.185 Uterus y = 0.0053x 2 0.4071x + 8.439 0.989 0.064 0.011 Skin y = 0.008x 2 0.5648x + 10.968 0.843 0.210 0.010 0.00 0.50 1.00 1.50 2.00 2.50 15.00 20.00 25.00 30.00 35.00 40.00 Organ Dose Conversion Coefficient Effective Diameter (cm) Thyroid Breast Lung Liver Stomach SI Colon Ovary Uterus Poly. (Thyroid) Poly. (Breast) Poly. (Lung) Poly. (Liver) Poly. (Stomach) Poly. (SI) Poly. (Colon) Poly. (Ovary) Poly. (Uterus)

PAGE 153

153 Table 8 40. Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI conversion coefficients for a chest protocol. Organ Polynomial Fit Equation R 2 Standard Error P value Thyroid y = 0.003x 2 0.1231x + 2.2618 0.089 0.553 0.830 Breast y = 0.0113x 2 0.7084x + 12.205 0.776 0.174 0.024 Lung y = 0.0106x 2 0.6475x + 10.862 0.811 0.120 0.016 Liver y = 0.0165x 2 0.9966x + 15.983 0.875 0.126 0.006 Stomach y = 0.0103x 2 0.6358x + 10.676 0.807 0.122 0.016 Skin y = 0.0086x 2 0.5363x + 9.662 0.742 0.135 0.034 Table 8 41. Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI conversion coefficients for an abdomen protocol. Organ Polynomial Fit Equation R 2 Standard Error P value Breast y = 0.0027x 2 + 0.0461x + 1.7383 0.879 0.168 0.005 Lung y = 0.0034x 2 + 0.1702x 1.5646 0.264 0.194 0.465 Lower Lung y = 0.0055x 2 + 0.2785x 2.6463 0.238 0.295 0.508 Liver y = 0.0012x 2 + 0.0119x + 1.6736 0.929 0.066 0.001 Stomach y = 0.0032x 2 + 0.1335x 0.0417 0.853 0.091 0.008 SI y = 0.0293x 2 1.6879x + 24.857 0.398 0.462 0.281 Colon y = 0.0054x 2 + 0.2564x 1.7872 0.729 0.139 0.038 Skin y = 0.0101x 2 0.6357x + 11.147 0.848 0.125 0.009 Table 8 42. Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI conversion coefficients for a pelvis protocol. Organ Polynomial Fit Equation R 2 Standard Error P value SI y = 0.019x 2 + 1.0105x 12.152 0.935 0.134 0.001 Colon y = 0.0029x 2 0.2084x + 4.4031 0.351 0.261 0.339 Ovary y = 0.003x 2 0.2x + 4.3086 0.430 0.133 0.246 Uterus y = 0.0037x 2 0.3002x + 6.6774 0.949 0.100 0.051 Skin y = 0.0073x 2 0.4943x + 9.5125 0.833 0.147 0.011

PAGE 154

154 Table 8 43. Polynomial fit equations for correlations between the effective diameter of the central slice and organ dose to CTDI conversion coefficients for a three phase liver protocol. Organ Polynomial Fit Equation R 2 Standard Error P value Breast y = 0.0324x 2 2.1117x + 35.47 0.729 0.764 0.038 Lung y = 0.0187x 2 1.1908x + 19.849 0.371 0.802 0.314 Lower Lung y = 0.0142x 2 0.9945x + 18.75 0.345 1.173 0.347 Liver y = 0.0248x 2 1.6439x + 29.132 0.900 0.353 0.003 Stomach y = 0.0224x 2 1.4931x + 27.06 0.958 0.212 0.000 SI y = 0.0757x 2 4.4089x + 65.224 0.540 1.130 0.143 Colon y = 0.0244x 2 1.5902x + 28.107 0.833 0.416 0.011 Skin y = 0.0356x 2 2.1848x + 36.19 0.791 0.435 0.020 Table 8 44. Calculated organ doses vs. measured organ doses for a trauma protocol and an overweight subject. CADAVER 5 Organ CAP_CTDI Dose (mGy) AP_CTDI Dose (mGy) Calculated Dose (mGy) Measured Dose (mGy) % Difference Breast 19.3 20.5 39.8 38.7 2.9 Lung 14.4 15.2 29.6 29.4 0.8 Liver 18.6 19.0 37.6 35.4 6.0 Stomach 20.6 22.1 42.7 40.5 5.5 SI 22.2 23.0 45.2 47.8 5.3 Colon 20.5 21.5 42.0 49.6 15.2 Ovary 18.2 19.6 37.8 40.8 7.3 Uterus 18.2 18.1 36.3 40.6 10.6 Skin 21.6 22.0 43.7 51.4 14.9

PAGE 155

155 Table 8 45. Calculated organ doses vs. measured organ doses for a trauma protocol and an obese subject. CADAVER 7 Organ CAP_CTDI Dose (mGy) AP_CTDI Dose (mGy) Calculated Dose (mGy) Measured Dose (mGy) % Difference Breast 22.5 21.4 43.9 47.0 6.6 Lung 18.9 20.2 39.1 38.3 1.9 Liver 20.1 19.2 39.3 38.8 1.3 Stomach 23.0 20.5 43.5 50.6 14.1 SI 23.1 20.5 43.5 44.5 2.1 Colon 22.8 21.1 43.8 40.3 8.6 Ovary 20.9 18.8 39.7 40.8 2.6 Uterus 18.4 17.4 35.8 34.8 3.1 Skin 24.4 24.5 48.9 52.1 6.2 Table 8 46. Calculated organ doses from CAP equations vs. measured organ doses for a chest protocol and a small, medium, and large subject. Organ Measured Dose (mGy) Calculated Dose (mGy) Dose Diff (mGy) % Diff CAD 4 Breast 8.5 8.9 0.4 4.4 Lung 8.7 7.7 1.1 12.4 Liver 9.9 9.7 0.2 2.4 Stomach 8.2 8.6 0.4 4.4 Skin 9.9 11.7 1.8 18.2 CAD 5 Thyroid 22.1 12.7 9.5 42.7 Breast 15.1 16.5 1.4 9.4 Lung 14.9 12.5 2.4 16.3 Liver 12.5 16.1 3.6 28.9 Stomach 13.4 17.5 4.1 30.3 Skin 18.9 18.9 0.1 0.4 CAD 6 Thyroid 22.3 16.4 5.9 26.6 Breast 16.8 15.9 0.9 5.4 Lung 15.0 12.5 2.5 16.9 Liver 16.0 14.3 1.7 10.8 Stomach 12.3 16.9 4.6 37.3 Skin 20.7 17.0 3.7 17.8

PAGE 156

156 Table 8 47. Calculated organ doses from CAP equations vs. measured organ doses for an abdomen protocol and a small, medium, and large subject. Organ Measured Dose (mGy) Calculated Dose (mGy) Dose Diff (mGy) % Diff CAD 4 Breast 9.5 12.0 2.5 26.7 Lung 5.6 10.5 4.8 86.4 Lower Lung 8.7 10.5 1.7 19.9 Liver 9.2 13.1 3.9 42.6 Stomach 9.7 11.4 1.7 17.8 SI 13.9 14.0 0.1 0.7 Colon 8.4 12.8 4.4 52.9 Skin 13.4 16.0 2.5 18.8 CAD 5 Breast 20.8 25.4 4.6 22.1 Lung 10.3 18.9 8.6 84.0 Lower Lung 16.8 18.9 2.1 12.6 Liver 19.7 24.4 4.7 24.1 Stomach 22.1 27.0 5.0 22.4 SI 7.8 29.2 21.3 272.6 Colon 21.9 26.9 5.0 22.8 Skin 26.0 28.4 2.4 9.2 CAD 6 Breast 12.6 23.7 11.2 88.8 Lung 8.2 20.2 12.0 146.6 Lower Lung 11.9 20.2 8.3 69.5 Liver 20.1 21.2 1.1 5.3 Stomach 22.4 24.1 1.7 7.4 SI 25.6 24.1 1.4 5.6 Colon 20.4 23.9 3.5 17.1 Skin 27.8 25.9 1.9 6.9

PAGE 157

157 Table 8 48. Calculated organ doses from CAP equations vs. measured organ doses for a pelvis protocol and a small, medium, and large subject. Organ Measured Dose (mGy) Calculated Dose (mGy) Dose Diff (mGy) % Diff CAD 4 SI 9.3 14.9 5.7 61.5 Colon 11.2 13.7 2.5 22.1 Ovary 11.8 11.0 0.8 6.6 Uterus 12.9 13.6 0.7 5.7 Skin 16.6 16.0 0.6 3.6 CAD 5 SI 26.4 28.0 1.6 6.0 Colon 13.8 26.2 12.4 90.0 Ovary 22.1 23.9 1.8 8.0 Uterus 24.4 22.0 2.4 9.8 Skin 27.6 26.8 0.8 2.9 CAD 6 SI 5.8 23.0 17.2 294.8 Colon 14.0 23.3 9.3 66.5 Ovary 24.3 21.1 3.2 13.3 Uterus 19.3 19.1 0.2 1.1 Skin 26.6 26.3 0.3 1.3

PAGE 158

158 CHAPTER 9 DIS CUSSION 9.1 Organ Doses from Helical vs. Ultra Helical A cquisitions 9.1.1 Over Ranging Effects on Organ D ose For every out of field organ analyzed, the most dramatic increase in dose was observed for the widest ultra helical detector configuration, 0.5 mm x 160. The dose increase is due to the increase in scanning length. This concept is illustrated in Figures 8 1 through 8 3 for the anatomical ranges of interest that were scanned in this study. When the widest detector configuration is used, the irradiated length encompasses some organs that may have otherwise only been exposed to scatter radiation. However, for some organs the variation in dose is more subtle. This result is owed to anatomical variations between subjects. For instance, with the abdomen protocol shown in Figure 8 3, the dose to the uterus of cadaver 4 appears almost as a horizontal line wit h minimal deviation. This organ was located farther away from the primary beam than the other organs shown. Hence, the wider beam widths did not result in a large difference in dose. The opposite effect is observed with the small intestine doses of subj ects 6 and 7 for the chest protocol. The effect of over ranging is more dramatically displayed in Figure 8 2, where differences of almost 10 mGy are observed from the 0.5 mm x 64 detector configuration to the 0.5 mm x 160 configuration. With the head p rotocol, F igure 8 1 illustrates a large discrepancy in dose for cadavers 6 and 7 when utilizing the 0.5 mm x 160 detector configuration as compared to the 0.5 mm x 32 detector configuration. The thyroid gland on both of these subjects was located 3 4 cm f rom the end of the planned scan range, i.e., the base of the skull. In examining the over ranging lengths, it is realized that for the 0.5 mm x 160 beam

PAGE 159

159 width, the additional length of the scan is between 5 to 7 cm at both the start and end of the scan. Thus, the increase in dose for the largest detector configuration is attributed to the effect of over ranging. Conversely, the same dose effect is not detected with cadaver 5, even though the calculation of over ranging length results in the same exact va lues as with cadavers 6 and 7. The source of this discrepancy is the use of a tilt for this head protocol. When gantry tilt is used with head exams, the gantry angle is aligned with the infraorbital margin, and for this exam that alignment resulted in a 9 degree tilt. Thus, the tilt resulted in an increase in distance between the primary beam and the thyroid, and counteracted the effect of over ranging. When optimizing or designing CT scanning protocols, the effect of over ranging needs to be ta ken into account in the dose analysis. For instance, for trauma protocols it might be advantageous to use the largest ultra helical acquisition in order to achieve the fastest possible scan. In addition, for trauma protocols all radiosensitive organs are likely to be within the range of the primary beam and no consideration to radiosensitive organs just outside the scan range needs to be given. However, for pediatric protocols, careful attention is often paid to the scan range in order to reduce doses to radiosensitive organs. In these cases, a standard helical detector configuration, such as 0.5 x 64 mm might be the most adequate mode of acquisition. This would still allow for a fast scan, in order to prevent motion artifacts, while minimizing the dose to radiosensitive organs outside of the scan field of view. 9.1.2 Primar y Beam Organ Dose Differences between Helical and Ultra Helical A cquisitions In Section 8.4.2.1, organ dose results were presented for four different detector configurations and fou r different cadaveric subjects. The minimum average organ dose

PAGE 160

160 for all 7 protocols was observed with the smallest detector configuration, 0.5mm x 64. The maximum doses measured for all 7 protocols were a result of scans performed at 0.5 mm x 100 and 0.5 mm x 160 detector configurations. Figures 9 1 through 9 3 reaffirm the observed trend of increasing organ doses with increasing beam widths. While the geometric efficiency is improved with increasing beam widths, the utilization of tube current modulati on with all body protocols proves to be the dominant factor when determining organ doses. When the tube current modulation algorithm, Sure Exposure 3D (Toshiba America Medical Systems, Tustin, CA), plans out the tube current for each rotation or half rota tion, an optimal mA is chosen based on the area covered by that rotation on the patient. When wider beam widths are employed, the area the algorithm is averaging over is increased. For example, with a chest protocol when the beam width changes from 32 mm to 80 mm, a rotation that would have only included lungs in the mA calculation might now include part of the liver, a higher attenuating organ, which would require the mA for that entire rotation to be higher. Essentially, as the beam width increases, th e effectiveness of the tube current modulation algorithm decreases, and allows for less variation in mA for each rotation. This concept is illustrated in Figures 8 5 through 8 7 for the anatomical ranges of interest that were scanned in this study. The 0.5 mm x 80 configuration is closest in size to that of the 0.5 mm x 64 configuration. It is also the closest in average organ dose values to that of the 64 slice configuration. The 0.5 mm x 160 configuration presents the largest deviation in size and av erage organ doses from the 0.5 mm x 64 configuration. From Table 8 22

PAGE 161

161 average organ doses from the chest protocol increase by 29 percent for the 0.5 mm x 160 detector configuration. While the organ dose results presented in 8.4.2.1 were intended to only i nclude organs that remained in the primary scan range, due to anatomical variations among subjects, some organs remained outside the selected scan range for only 1 or 2 subjects. The effect of over ranging is again observed for the breast dose when the ab domen and three phase liver protocols were used with cadaver 7. For example, the smallest average organ dose of 15.3 mGy was measured for the breast of cadaver 7 and 64 slice selection, where over ranging has a minimal effect. When the 160 slice configur ation is used, the same breast dose increases to 48.4 mGy, now within the primary beam due to over ranging. This is shown in Figure 9 3. These conclusions are in accordance with those discussed in 9.1.1. 9.2 Organ Doses Head Protocols As described in Sectio n s 6.2 and 6.3 for the head and CTA head protocols, the detector configurations that are currently used in clinical prac tice only include 0.5 mm x 32 and 0.5 mm x 64. For the brain perfusion protocol, volumetric acquisitions (i.e. 0.5 mm x 320) are also c linically relevant. The average organ doses for a standard head protocol at SUF for the brain, lens and skin were 36.2 mGy, 55.6 mGy and 56 mGy respectively Tan et al. used 3 different cadaveric heads filled with gauze to measure lens dose from a helic al head protocol on a Siemens 64 slice MDCT (Siemens Medical Solutions, Forchheim, Germany). 90 The average dose observed in that study was 47.2 mGy, which is very close to the average lens dose of 50.3 mGy measured in this research for the 0.5 mm x 64 det ector configuration. 90 For the standard CTA head

PAGE 162

162 protocol at Shands, the average organ doses were 107.4 mGy, 16.1 mGy, 158 mGy, and 160.1 mGy for the brain, thyroid, lens, and skin respectively The highest head organ doses were observed with the brain perfusion protocol, and the average organ doses for the brain, lens, and skin were 169 mGy, 299 mGy, and 298 mGy respectively Employing a bismuth shield over the eyes can reduce the lens dose by 20 24 %. Alternate studies have shown a similar percent dose reduction with the use of a lens bismuth shield. Perisinakis et al. studied eye lens dose reduction with the use of a bismuth shield for pediatric multi phase head studies. The study reports that with the use of an anthropomorphic phantom and thermo luminescent dosimeters, a dose reduction of 34 % was observed when t he bismuth shield was utilized. 102 In an alternate study conducted at the Mayo Clinic (Rochester, MN), lens dose reduction techniques were examined on a Siemens Definition Flash with an anthropomorphic phantom and OSLDs. A dose re duction of 26.4 % was observed. 103 Figure 8 1 shows all organ doses measured for a standard head protocol at various detector configurations for organs that are both in the primary scan range and those that are just outside of the scan range and subject to the effects of over ranging. When the beam width incr eases, the two sets of organs have opposite responses. For organs that are included in the primary scan range, a decrease in dose is observed with an increase in beam width or detector configuration. This decrease in dose can be attributed to the increa sed geometric efficiency described in 8.2.2. In addition, all head scans are performed with a fixed tube current and no use of tube current modulation. This contributes to the clear relationship observed between dose

PAGE 163

163 and geometric efficiency. Furthermor e, a similar trend occurs with the CTDI Vol of each scan, shown as the black line in Figure 8 1. The opposite trend is observed for organs just outside of the primary scan range. The thyroid dose increases with increasing beam width. As discussed in Sect ion 8.2.3, this dose increase is accredited to the increase in scan length caused by over ranging. The head organ doses that are presented with this research are all in accordance with those observed in the literature and remain under current thresh old s for deterministic effects. 15 However, the use of these protocols should adhere to appropriateness criteria, and should always be optimized for the diagnostic task at hand. 9.3 Fetal Dose Estimates Direct measurement of in utero doses has been accomplish ed with this research. This methodology can be used to estimate 1 st trimester fetal doses from three different clinically relevant protocols, including pulmonary embolism, chest abdomen pelvis, and trauma protocol s The dose to the uterus from the pulmona ry embolism protocol is only a result of scatter ed radiation. It was observed for all subjects that the fetal dose estimate for this protocol was under 1 mGy. All additional organs located completely outside of the scanning length, which includes the con tribution from over ranging, are also only exposed to scatter and their corresponding doses remain under 1 2 mGy. The chest abdomen pelvis and trauma protocol s do result in primary radiation covering the pelvis. Th e average dose to the uterus from a chest abdomen pelvis protocol is 19.4 mGy. For a trauma protocol, the average uterus dose is 34.6 mGy. The body portion of the trauma protocol does result in higher doses when compared to

PAGE 164

164 just a CAP because this protocol consists of a CAP and an AP scan or 2 p asses over the pelvic region, compared to only one in the standard CAP protocol. These directly measured values can be compared with existing fetal dose estimates published elsewhere In a study by Hurwitz et al. an adult female CIRS phantom and thermolu minescent dosimeters we re used to measure doses to the uterus for a pulmonary embolism protocol with a resulting fetal dose of 0.32 mGy. 97 This value matches the uterus dose measured for 2 of the subjects in this study Angel et al. created 24 voxel mode ls of pregnant patients and simulated an abdo men pelvis scan MCNPX, a radiation transport code. The resulting fetal dose estimates ranged from 16 mGy to 31 mGy. 98 Comparatively, these estimates were slightly higher than results of this study for a CAP pr otocol. However, the voxel phantoms developed by Angel et al. were at an average gestat ional age of 20 weeks, while the estimates of the study presented here are only valid for the first trimester. In a study by Jaffe et al., MOSFETS (metal oxide semicond uctor field effect transistors) were used to measure uterine doses in an adult female CIRS phantom. 95 These methods resulted in a dose of 14.3 mGy for CAP pro tocol, very close to the dose this study measured of 14.1 m Gy for the smallest cadaveric subject and a CAP protocol. When located in the primary scan range, as with the CAP and trauma protocols, fetal doses remain under 50 mGy. Hence, this research proves that these doses will result in a negligible increased risk to the fetus as a result of the radiation exposure from such CT examinations Finally, when the clinical indications for these exams are met, they should be performed on pregnant patients with a full understanding of the benefits and risks or rather, minimal risks to the fetus. Additi onally, this data can serve to

PAGE 165

165 validate past and future studies of fetal doses that result from CT exams, as this is an ongoing clinical issue. 9.4 S ize S pecific D ose E stimate Comparison It is observed with this data that the SSDE does accurately describe the average dose absorbed in the body. Across the board, comparisons with the average of all organ doses and the central slice SSDE yielded the most similar results. The percent difference between those two values ranges from 0.4 to 21.8 percent. The farthe st deviations between measured organ doses and the SSDE calculated for each organ location existed for the thyroid, breasts, lungs, ovaries, and skin. Most of these organs were located on the periphery of the patient where deviations would be expected due to to be an estimate of organ dose but rather an estimate of the average dose distributed in a patient. 52 Therefore, it makes sense that the SSDE would most accurately c ompare to the average of all organ doses as opposed to individual organ dose values for each subject. In conclusion, the SSDE can effectively be used as a patient dose descriptor in the clinic, while organ dose libraries and software that allow for organ dose calculation are still being developed. 9.5 Size Parameter Correlations The first relationship examined for the size parameter organ dose correlations was that between the effective diameter and organ doses for a CAP protocol. From Figure 8 9 it can be di scerned that as the effective diameter of a subject increases, so do the organ doses, but this relation is only true for effective diameters under 30 cm. When the effective diameter increases past 30 cm, the curve exhibits a downward tendency, indicating that any further increase in effective diameter results in a decrease

PAGE 166

166 subcutaneous soft tissue surrounding the internal organs, which can result in lower organ doses. Th is effect is only realized in larger patients where the tube current modulation effectively reaches its upper limit and is no longer useful. A set of equations to calculate organ doses from the knowledge of the scan CTDI Vol and central effective diameter of the patient are presented in Tables 8 39 through 8 43 for a CAP, chest, abdomen, pelvis, and three phase liver protocols. To evaluate the use of the organ dose equations in a clinical setting, a trauma protocol was used as a test case. The comparison between calculated organ doses from the equations of Table 8 39 and those measured in Table 8 22 for a trauma protocol were shown in Tables 8 44 and 8 45. For the two subjects, a maximum of 15 % difference was observed. The average percent difference for the large subject was 7.6 %, while the average percent difference for the obese subject was 5.2 %. Hence, these equations can reliably be utilized to estimate organ doses for additional protocols as long as the measurement protocol is followed properly. The second application explored with the protocol equation sets was if the CAP equations could be used for all protocols. The results for the comparisons of measured organ doses and calculated organ doses from the CAP equations are shown in Tables 8 46, 8 47, and 8 48 for the chest, abdomen, and pelvis protocols, respectively. The use of the CAP equations for the individual protocols relies on the assumption that all organs of concern are completely located within the scan range and hence are fully irradi anatomical organ distribution. The use of just the CAP protocol might be an overly

PAGE 167

167 conservative approach to estimating organ doses. Hence, all protocol equation sets will continue t o be used for organ dose estimates. For some organs and protocols, doses cannot be accurately calculated with these equations. Most organs with weak correlations are those that may or may not be included in the primary scan range depending upon anatomi cal variations among subjects. Correlations for those such organs prove to be extremely difficult to generate. That being said, a substantial amount of useful organ dose data can originate from the equations presented with this research, these correlatio ns will only strengthen with the Vol and the measurement of 2 dimensions on that scan, patient specific organ doses can be calculated for an array of CT protocols.

PAGE 168

168 Figure 9 1. Average organ doses in the primary scan range for the chest protocol at various detector configurations. Figure 9 2. Average organ doses in the primary scan range for the abdomen protocol at various detector configurations. 0.0 5.0 10.0 15.0 20.0 25.0 64 80 100 160 Average Organ Dose (mGy) # of Active Detector Channels Breast CAD5 Stomach CAD5 Liver CAD5 Lung CAD5 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 64 80 100 160 Average Organ Dose (mGy) # of Active Detector Channels Liver CAD6 SI CAD6 Skin CAD6

PAGE 169

169 Figure 9 3. Average organ doses in the primary scan range for the three phase liver protocol at various detector configurations. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 64 80 100 160 Average Organ Dose (mGy) # of Active Detector Channels Breast CAD7 Liver CAD7 Stomach CAD7

PAGE 170

170 CHAPTER 10 CONCLUSIONS 10.1 Summary The ultimate goals of this research project were: first, to develop sets of empirical equations that can be utilized to calculate patient specific organ doses for a group of commonly performed CT exams, and second, to investigate the effects of the ultra h elical acquisition mode on organ doses. Both of these goals were accomplished with the use of a standardized and reproducible direct organ dose measurement methodology utilizing commercially available optically stimulated luminescent dosimeters and perfor ming the measurements on cadaveric subjects in lieu of actual human patients. The organ dose equations put forth in this document were developed by utilizing data obtained from direct organ dose measurements performed with a series of cadaveric subjects a nd correlating the organ dose data with patient specific parameters for a group of clinically relevant CT protocols. The origin of these equations from the direct measurement of organ doses sets them apart from other methods of organ dose estimation that currently exist. The patient dose determination protocol presented in this document has immediate clinical implementation and is the first of its kind in the field. Physicists at SUF Radiology will have now the means to accurately determine individual pa tient doses from CT studies upon request in a matter of minutes. This information will be available to clinicians and patients and will be of great value in the clinical assessment of specific cases, such as the management of pregnant patients (based on f etal doses), IR patients at risk of high skin dose and sentinel events, and others. The ability to generate patient doses from individual CT scans will be used to

PAGE 171

171 initiate the collection, maintenance, and tracking of realistic lifetime organ doses in the EMR for all patients. For the second major goal, a total of four cadaveric subjects were utilized for the evaluation of the ultra helical acquisition mode. The analysis of organ dose variation for ultra helical acquisitions was divided into two categories : the first examined organ doses outside of the prescribed scanning range (PSR) of an exam, and the second examined organ doses within the PSR. The exposure to organs outside of the PSR is solely the result of beam over ranging. Beam over ranging increas es doses to organs outside of the prescribed scan range, and becomes more prominent with wider beam widths. For body scans, organ doses within the PSR show an overall increase in dose with an increase in the size of the detector configuration. Currentl y, all body scan protocols utilize tube current modulation, which becomes less effective as the beam width increases. The opposite effect was observed for head protocols, as most head exams are performed using fixed tube current, allowing for the geometri c efficiency of wider beams to result in a decrease in organ doses. Because ultra helical mode provides shorter scanning times which are critical in many clinical indications, it is likely to be used in an expanded role in CT studies. Knowledge of the c orresponding organ doses will be of significant clinical value as current protocols shift from 64 slice helical acquisitions to ultra helical acquisitions, aiding the SUF radiology practice committee in the decision process of protocol optimization. 10.2 Futu re Work Although, these equations do represent a great stride in the ability to determine patient specific organ doses, there are some limitations associated with their use. First, the work presented here is only associated with scanner from one manufactu rer. The

PAGE 172

172 use of CTDI Vol as a normalization factor may overcome this limitation and could lead to scanner independent estimates of organ dose, but the validity of this must be independently and thoroughly investigated. Second, while a great deal of time a nd effort went into the measurement of organ doses for 8 different subjects, it is still a relatively small sample size in terms of the actual patient population. Stronger correlations and smaller values of standard error could result from the inclusion o f additional subjects. Thus, future work to include additional subjects and various scanners would prove to be beneficial. While the organ doses measured in this research are representative of technology that is the current clinical standard, new technolo gical advances in CT continue to occur at a rapid pace. The use of iterative reconstruction algorithms instead of traditional filtered backprojection is the newest technological advancement in CT for which an evaluation of dose is needed. Many manufactur ers quote significant dose reductions with the use of iterative reconstruction. These claims need to be supported with actual organ dose measurements. 10.3 Final Words In conclusion, the organ doses and organ dose equations presented in this document span a wide range of practical applications. The organ doses included are the first directly measured doses in CT and account for new advancements in technology, such as the evaluation of the new ultra helical acquisition mode. In addition, the range of patien t sizes for which these organ doses are applicable, was expanded to include obese subjects, making the organ doses more representative of the changing patient population in the U.S.. The organ dose equations developed from these direct organ dose measurem ents are also both novel and widely applicable. In

PAGE 173

173 utilizing CTDI Vol as a dose normalization factor, the equations developed can be used for a broad array of protocols, and potentially, other scanners. In utilizing the effective diameter as a patient spe cific size parameter, the equations can be used for an extended range of patient sizes. These equations represent the first step in acquiring the ability to calculate patient specific organ doses in CT, which will in turn, result in accurate estimates of risk for this life saving diagnostic modality.

PAGE 174

174 LIST OF REFERENCES 1. Hounsfield GN. Computerized Transverse Axial Scanning (Tomography): Part 1. Description of System. British Journal of Radiology. 1973; 46:1016 1022. 2. Modality Benchmark Reports. http://www.imvinfo.com 2007. (Accessed March, 2012, at http://www.imvinfo.com/user/documents/content_documents/ nwsrad/MSC TDSandTOC.pdf 3. Modality Benchmark Reports. http://www.imvinfo.com 2012. (Accessed July, 2012, at http://www.imvinfo.com/user/documents/content_documents/ defdis/20120605045416458IMV2012CTPressRelease.pdf 4. Brenner DJ, Hall EJ. Computed tomography -an increasing source of radiation exposure. N Engl J Med 2007; 357:2277 84. 5. Bogdanich W. Radiation Overdo ses Point Up Dangers of CT Scans. New York Times 2009;Sect. A13. 6. National Council on Radiation Protection and Measurements. Ionizing Radiation Exposure of the Population of the United States, Report 160. 2009. 7. National Research Council of the National Acad emies. Health risks from exposure to low levels of ionizing radiation: BEIR VII, Phase 2. Washington, DC: The National Academies Press. 2006. 8. Siegel JA and Stabin MG. Radar Commentary: Use of Linear No Threshold Hypothesis in Radiation Protection Regulatio n in the United States. Health Phys. 2012; 102(1):90 99. 9. Separating Fact from Fantasy. Radiology. August 2012; 264(1):312 321. 10. Pearce MS, Salotti JA, Little MP, McHugh K, Lee C, Kim KP, Howe NL, Ronckers CM, Rajaraman P, Craft AW, Parker L, de Gonzalez AB. Radiation Exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study. Lancet. 2012; 380:499 505. 11. National Research Council of the Na tional Academies. Tracking radiation exposure from medical diagnostic procedures. Washington, DC: The National Academies Press. 2012. 12. The Joint Commission. Sentinel Event Alert Issue 47. August 24, 2011. ( www. jointcommission.org )

PAGE 175

175 13. International Commission on Radiation Protection. 2012 ICRP Statement on Tissue Reactions / Early and Late Effects of Radiation in Normal Tissues and Organs Threshold Doses for Tissue Reactions in a Radiation Protection Context. ICRP Publication 118. Ann. ICRP 41(1/2). 14. AMERICAN ASSOCIATION OF PHYSICISTS IN MEDICINE, Comprehensive 15. Methodology for the Evaluation of Radiation Dose in X Ray Computed Tomography, AAPM Rep. 111 New York 2010. 16. International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann. ICRP 37 (2 4). 2007. 17. International Commission on Radiological Protection. Managing Patient Dose in Multi Detector Computed Tomograph y (MDCT). ICRP Publication 102. Ann. ICRP 37 (1). 2007. 18. Bushberg JT, Siebert JA, Leidholdt Jr EM, Boone JM. The Essential Physics of Medical Imaging. Second ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2002. 19. Hurlock GS, Higashino H, Mochizuki T. Hi story of cardiac computed tomography: single to 320 detector row multislice computed tomography Int J Cardiovasc Imaging 2009; 25:31 42. 20. McCollough CH, Zink FE Med Phys 1999; 26:2223 2230. 21. Bongartz G Golding SJ, Jurik AG, Leonardi M, van Persijn E, van Meerten, Rodrguez R, Schneider K, Calzado A, Geleijns J, Jessen KA, Panzer W, Shrimpton PC, Tosi G. European Guidelines for Multislice Computed Tomography Funded by the European Commission Contract number FIGM CT2000 20078 CT TIP March 2004 22. Rogalla P, Kloeters C, Hein PA. CT Technology Overview: 64 slice and Beyond. Radiol Clin N Am 2009; 47:1 11. 23. IEC, 2002. Medical Electrical Equipment. Part 2 44: Particular requirements for the safety of x ray eq uipment for computed tomography. IEC publication No. 60601 2 44. Ed. 2.1. International Electrotechnical Commission (IEC) Central Office, Geneva, Switzerland. 24. Weigold W, Olszewski M, Walker M. Low dose prospectively gated 256 slice coronary computed tomogr aphic angiography. Int J Cardiovasc Imaging. 2009; 25 :2 17 230.

PAGE 176

176 25. Schilham A, Van der Molen AJ, Prokop M, De Jong HW. Overranging at Multi section CT: An Underestimated Source of Excess Radiation Exposure. Radiographics 2010; 30: 1057 1067. 26. Shrimpton PC, Hillier MC, Lewis MA, Dunn M. Doses from computed tomography (CT) examinations in the UK: 2003 review Chilton, UK: National Radiological Protection Board, 2005: report NRPB W670. 27. Kopka, L, Funke M, Breiter N et al., 1995. Anatomically adapte d CT tube current: Dose reduction and image quality in phantom and patien t studies. Radiology 197: 292 28. Matsubara K, Koshida K, Ichikawa K, Suzuki M, Takata T, Yamamoto T, Matsui O. Misoperation of CT automatic tube current modulation systems with inappropr iate patient centering: phantom studies. AJR Am J Roentgenol. 2009; 192 :862 865. 29. Duan X, Wang J, Christner JA, Leng S, Grant KL, McCollough CH. Dose reduction to anterior surfaces with organ based tube current modulation: evaluation of performance in a phantom study. AJR Am J Roentgenol 2011 Sep; 197(3):689 95. 30. Hausleiter J, Meye r T, Hadamitzky M et al. Radiation dose estimates from cardiac multislice computed tomography in daily p ractice: impact of different scanning protocols on effective do se estimates. Circulation. 2006; 113(10): 1305 1310. 31. Kang EJ, Lee KN, Kim DW, Kim BS, Choi S, Park BH, Oh JY. Triple rule out acute c hest pain evaluation using a 320 row detector volume CT: a comparison of the wide volume and helical modes. Int J Cardiovasc Imaging 2012 Jun; 28 Suppl 1:7 13 32. Van der Molen AJ, Geleijns J. Overranging in multi section CT: quantification and relative contribution to dose comparison of four 16 section CT scan ners. Radiology. 2007; 242(1):208 216. 33. Christner JA, Zavaletta VA, Eusemann CD, Walz Flannigan AI, McCollough CH. Dose Reduction in Helical CT: Dynamically Adjustable z Axis X Ray Beam Collimation. AJR. 2010; 194:W49 W55. 34. Fleischmann D and Boas FE. Computed tomography old ideas and new technology. Eur Radiol. 2011; 21:510 517. 35. McCollough CH, Bruesewitz MR, Kofler JM. CT Dose Reduction and Dose Management Tools: Overview of Available Options. Radio logy. March 2006; 26:503 512.

PAGE 177

177 36. Deak PD, Langner O, Lell M, Kalendar WA. Effects of Adaptive Section Collimation on Patient Radiation Dose in Multisection Spiral CT. Radiology. July 2009; 252: 140 147. 37. Shope, T., R. Gagne, and G. Johnson ibing the doses delivered by transmission x Med Phys. 1981; 8:488 495. 38. U. S. Food and Drug Administration (FDA). Diagnostic X Ray Systems and Their Major Components. Code of Federal Regulations 21 CFR 1020.33, 1984. 39. International Electrotechnical Commission. Particular requirements for the safety of x ray equipment for computed tomography. Ed2. 60601 2 44. 2001. 40. American Association of Physicists in Medicine. Report 96. The Measurement, Reporting, and Management of R adiation Dose in CT. College Park, MD. 2007. 41. Jucius, RA and Kambic, GX. Radiation dosimetry in computed tomography. Application of optical instrumentation in medicine Part VI. Proceedings of the Society of Photo Optical Instrumentation in Engineering.1977; 127: 286 295. 42. Suzuki A and Suzuki MN. Use of a pencil shaped ionization chamber for measurement of exposure resulting from a computed tomography scan. Med. Phys. 1978; 5 :536 539. 43. International Electrotechnical Commission (IEC). Medical Electrical Equipmen t. Part 2 44: Particular requirements for the safety of x ray equipment for computed tomography. IEC publication No. 60601 2 44. Ed. 2.1: International Electrotechnical Commission (IEC) Central Office: Geneva, Switzerland, 2002. 44. Dixon R L. A new look at CT dose measurement: Beyond CTDI Med. Phys. 2003; 30: 1272 80. 45. Boone JM. The trouble with CTD10 0. Med Phys. 2007; 34(4):1364 1371 46. S. Mori, M. Endo, K. Nishizawa, T. Tsunoo, T. Aoyama, H. Fujiwara, and K. Murase, Enlarged longitudinal dose profiles in cone b eam CT and the need for modified dosimetry. Med. Phys. 2005; 32 : 1061 1069. 47. S. Mori, K. Nishizawa, M. Ohno, and M. Endo, "Conversion factor to CT dosimetry to assess patient dose using a 256 slice CT scanner," Brit. J. Radiol. 2006; 79 : 888 892. 48. J. Geleijns A.M. Salvado, P.W. de Bruin, R. Mather, Y. Muramatsu, and M.F. Nitt Gray, Computed tomography dose assessment for a 160 mm wide, 320 detector row, cone beam CT scanner. Phys. Med. Biol. 2009; 54 :3141 3159.

PAGE 178

178 49. INTERNATIONAL ELECTROTECHNICAL COMMISSION, Medic al Electrical Equipment Part 2 44 Edition 3, Amendment 1: Particular requirements for basic safety and essential performance of X ray equipment for computed tomography, IEC 60601 2 44 Edition 3, Amendment 1; 62B/804/CD, COMMITTEE DRAFT (CD), IEC Geneva (2010). 50. International Atomic Energy Agency. Human Health Series Report No. 5 Status of Computed Tomography Dosimetry for Wide Cone Beam Scanners. IAEA Vienna, 2011. 51. C.H. McCollough, S. Leng, L. Yu, D.D. Cody, J.M. Boone, M.F. McNitt Dose Index a nd Patient Dose: They are Not the Same Thing, 259:311 316 52. Shrimpton P Reference doses for computed tomography. Radiological protection bulletin 193. Chilton, England: National Radiological Protection Board, 1997 ; 16 19. 53. American Association of Physicists in Medicine. Size Specific Dose Estimates (SSDE) in Pediatric and Adult Body CT Examinations. College Park, MD; 2011. 54. Jacobi, W. The Concept of Effective Dose A Proposal for the Combination of Organ Doses. Radiat. Environ. Bioph ysics. 1975; 12: 101 109. 55. International Committee on Radiological Protection. Recommendations of the International Committee on Radiological Protection. ICRP Publication 26. Ann. ICRP 1 (3). 1971. 56. European Guidelines on Quality Criteria for Computed Tomography (EUR 16262 EN, May 1999). Available at: www.drs.dk/guidelines/ct/quality/index.htm 57. Bongartz G, Golding SJ, Jurik AG, et al. E uropean guidelines for multislice computed tomography. Funded by the European Commission contract number FIGM CT2000 20078 CT TIP 2004. Luxembourg: European Commission 58. Shrimpton PC, Wall BF, Yoshizumi TT, Hurwitz LM, Goodman PC. Effective dose and dose l ength product in CT. Radiology 2009; 250:604 605 59. Kalender W. A. Schmidt B. Zankl M. Schmidt M. A PC program for estimating organ dose and effective dose values in computed tomography. Eur. Radiol.1999; 9:555 562. 60. expo A no vel program for dose evaluation in 2002; 174:1570. 61. Lee C, Williams JL, Bolch WE. The UF series of tomographic computational phantoms of pediatric patients. Med Phys 2005; 32:3537 48.

PAGE 179

179 62. Lee C, Lodwick D, Hur tado J, Pafundi D, Williams JL, Bolch WE. The UF Family of Reference Hybrid Phantoms for Computational Radiation Dosimetry. Phys Med Biol. January 21, 2010; 55(2):339 363. 63. Turner AC, et al., The feasibility of a scanner independent technique to estimate or gan dose from MDCT scans: using CTDI vol to account for differences between scanners. Med. Phys. 2010; 37:1816 1825 64. Huda W, Ogden KM, Lavallee RL et al. In patient to isocenter KERMA ratios in CT. Med Phys. 2011; 38:5362 5369 65. Strauss KJ. KERMA ratios vs SSDE: is one better at estimating pediatric CT radiation doses?. Pediatr Radiol. 2012 May; 42(5):525 6. 66. Strauss KJ and Goske MJ. Estimating Pediatric Radiation Dose dur ing CT. Pediatr Radio l. 2011 Sep; 41 Suppl 2:472 482. 67. Knoll GF. Radiation Detection and Measurement. Third ed. United States of America: John Wiley & Sons, Inc; 2000. 68. National Institute of Standards and Technology, "Ionizing radiation measurements: dosimetry of x ra y, gamma rays, and electrons," 2009 Available from: http://ts.nist.gov/MeasurementServices/Calibrations/x gamma ray.cfm 69. Markey BG, Colyott LE, McKeever SWS Time resolved optically stimulated Al 2 O 3 :C. Radiat Meas. 1995; 24:457 463. 70. Yukihara EG and McKeever SWS. Optically Stimulated Luminescence Fundamentals and Applications. John Wiley & Sons Ltd; 2011. 71. Landauer I. microStar Dosimeter: Th e Power of the Dot. In: http://www.landauer.com/uploadedFiles/Healthcare_and_Education/Solutions/na noDot%20Spec%20Sheet.pdf Glenwood, IL; 2 008. 72. Jursinic PA. Characterization of optically stimulated luminescent dosimeters, OSLDs, for clinical dosimetric measurements. Med Phys. 2007; 34:4594 604. 73. Lavoie LK. Organ dose measurements from multiple detector computed tomography using a commercial do simetry system and tomographic, physical phantoms. In. [Gainesville, Fla.]: University of Florida; 2009. 74. Al Senan RM and Hatab MR. Characteristics of an OSLD in the diagnostic energy range. Med Phys. July 2011; 38:4396 4405. 75. Lavoie L, Ghita M, Brateman L, Arreola M. Characterization of a Commercially Available Optically Stimulated Luminescent Dosimetry System for Use in Computed Tomography. Health P hysics. September 2011; 101(3): 299 310.

PAGE 180

180 76. Bos A J J. High sensitivity thermoluminescence dosimetry Nucl. Instrum Methods Phys Res B. 2001; 184: 3 28. 77. Yukihara EG, Ruan C, Gasparian PB, Clouse WJ, Kalavagunta C, Ahmad S. An optically stimulated luminescence system to measure dose profiles in x ray computed tomography. Phys Med Biol 2009; 54:6337 6352. 78. Wang J, Duan X, Christner JA, Leng S, Grant KL, McCollough CH. Bismuth Shielding, Organ based Tube Current Modulation, and Global Reduction of Tube Current for Dose Reduction to the Eye at Head CT. Radiology. January 2012; 262(1): 191 198. 79. Al Senan R, Muell er DL, Hatab MR. Estimating Thyroid Dose in Pediatric CT Exams from Surface Dose Measurement. Phys Med Biol. 2012; 57:4211 4221. 80. Griglock T. Determining Organ Doses from Computed Tomography Scanners Using Cadaveric Subjects. In. [Gainesville, Fla.]: Univer sity of Florida; 2012. 81. Jones AK, Hintenlang DE, Bolch WE. Tissue equivalent materials for construction of tomographic dosimetry phantoms in pediatric radiology. Med Phys 2003; 30:2072 81. 82. Winslow JF, Hyer DE, Fisher RF, Tien CJ, Hintenlang DE. Construction of anthropomorphic phantoms for use in dosimetry studies. J Appl Clin Med Phys 2009; 10:2986. 83. Cristy, M.; Eckerman, KF. Specific absorbed fractions of energy at various ages from internal photon sources. Oak Ridge National Laboratory; Oak Ridge, TN: 1987. ORNL TM 8381 84. Xu XG, Taranenko V, Zhang J, Shi C. A boundary representation method for designing whole body radiation dosimetry models: pregnant females at the ends of three gestational periods -RPI P3, P6 and P9. Phys Med Biol 2007; 52:7023 7044. 85. Segars WP, Mahesh M, Beck TJ, Frey EC, Tsui BM. Realistic CT simulation using the 4D XCAT phantom. Med Phys 2008; 35:3800 3808. 86. Germerott T, Preiss US, Ebert LC, Ruder TD, Ross S, Flach PM, Ampanozi G, Filograna L, Thali MJ. A New Approach in Virtopsy: Postmort em Ventilation in Multislice Computed Tomography. Legal Medicine. 2010; 12:276 279. 87. Levy AD, Harcke HT, Mallak CT. Postmortem Imaging MDCT Features of Postmortem Change and Decomposition. Am J Forensic Med Pathol. March 2010; 31(1):12 17.

PAGE 181

181 88. Christe A, Flach P, Ross S, Spendlove D, Bolliger S, Vock P, Thali MJ. Clinical Radiology and Postmortem Imaging (Virtopsy) are not the same: Specific and Unspecific Postmortem Signs. Legal Medicine. 2010; 12:215 222. 89. Chew FS, Relyea Chew A, Ochoa ER. Postmortem computed tomography of cadavers embalmed for use in teaching gross anatomy. J Comput Assist Tomogr 2006; 30:949 54. 90. Serhal CB, Jacobs R, Gijbel F, Bosmans H, Hermans R, Quirynen M, van Steenberghe D. Absorbed doses from spiral CT and conventional spiral tomography: a phantom vs. cadaver study. Clin Orgal Impl. Res. 2001; 12:473 478. 91. Tan JS, Tan KL, Lee JC, Wan CM, Leong JL, Chan LL. Comparison of eye lens dose on neuroimaging protocols between 16 and 64 section multidetector CT: achieving the lowest possible dose. AJNR Am J Neuroradiol. 2009; 30 (2):373 377. 92. International Commission on Radiological Protection. Biological Effects after Prenatal Irradiation (Embryo and Fetus). ICRP Publication 90 Ann ICRP 2003; 33:1 2. 93. American College of Radiology. ACR practice guide line for imaging pregnant or potentially preg nant adolescents and women with ionizing radiation. Reston, Va. 2008. 94. National Council on Radiation Protection and Measurements. Medical radiation exposure of pregnant and potentially pregnant women. NCRP repor t no. 54. Bethesda, Md: National Council on Radiation Protection and Measurements, 1977. 95. McCollough CH, Schueler BA, Atwell TD, et al. Radiation exposure and pregnancy: when should we be concerned? RadioGraphics. 2007; 27:909 918. 96. Jaffe TA Yoshizumi TT, Toncheva GI, Nguyen G, Hurwitz LM, Nelson RC. Early first trimester fetal radiation dose estimation in 16 and 64 MDCT scanners and low dose imaging protocols. AJR 2009; 193:1019 1024. 97. Damilakis J et al. A method of estimating conceptus doses resulting fro m multi detector CT examinations during all stages of gestation. Med Phys. 2010; 37(12):6411 6420. 98. Hurwitz LM, Yoshizumi T, Reiman RE, et al. Radiation dose to the fetus from body MDCT during early gestation AJR 2006; 186:871 876.

PAGE 182

182 99. Angel E, Wellnitz CV, G oodsitt MM, et al. Radiation dose to the fetus for pregnant patients undergoing multidetector CT imaging: Monte Carlo simulations estimating fetal dose for a range of gestational age and patient size. Radiology 2008; 249:220 27. 100. Boone JM, Geraghty EM, Seiber JA, Wootton Gorges SL. Dose reduction in pediatric CT: A rational approach. Radiology 2003; 228: 352 360. 101. Menke J. Comparison of different body size parameters for individual dose adaptation in body CT of adults. Radiology. 2005; 565 571. 102. World Healt h Organization. BMI Classification Chart. Available from: http://apps.who.int/bmi/index.jsp?introPage=intro_3.html Accessed July 2012. 103. Perisinakis K, Raissaki M, Theocharopoulos N, Damilakis J, Gourtsoyiannis N. Reduction of eye lens radiation dose by orbital bismuth shielding in pediatric patients undergoing CT of the head: a Monte Carlo study. Med Phys. 2005 Apr; 32(4):1024 30. 104. Wang J, Duan X, Christner JA, Leng S, Grant KL, McCollough CH. Bismuth shielding, organ based tube current modulation, and global reduction of tube curr ent for dose reduction to the eye at head CT. Radiology. 2012 Jan; 262(1):191 8.

PAGE 183

183 BIOGRAPHICAL SKETCH Lindsay Sinclair was born and raised in St. Petersburg, Florida She completed her undergraduate degree at the University of Florida earning a Bachelor of S cience in 2007 with a major in nuclear engineering sciences and a concentration in medical p degree in 2009 with a major in nuclear engineering sciences and a concentration in medical p hysics. She received her Ph.D. from the University of Florida in the summer of 2013. She curren tly lives in Po rtland, OR and works as a diagnostic imaging physicist at Oregon Health and Science University. She is looking forward to broadening her horizons, but will always be a Florida Gator at heart.