<%BANNER%>

A Fiber-Optic Coupled Point Dosimetry System for the Characterization of Multi-Detector Computed Tomography

Permanent Link: http://ufdc.ufl.edu/UFE0022737/00001

Material Information

Title: A Fiber-Optic Coupled Point Dosimetry System for the Characterization of Multi-Detector Computed Tomography
Physical Description: 1 online resource (103 p.)
Language: english
Creator: Moloney, William
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Nuclear and Radiological Engineering -- Dissertations, Academic -- UF
Genre: Nuclear Engineering Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The importance of radiation dose and the associated risk from x-ray computed tomography (CT) has been highlighted by the scientific community as recent developments in technology have greatly improved the capabilities of CT scanners. As image quality rapidly improves, it is necessary to quantify the radiation dose delivered from new imaging modalities, and evaluate the means for reducing patient dose to as low a level as possible. Developments in computed tomography technology have thus offered a change in the clinical use of CT, prompting the need to evaluate the radiation risk versus medical benefit and calling into question traditional dosimetry methods that may no longer be applicable to new scanner capabilities. The goal of this work was to develop a prototype point dosimeter for the characterization of the radiation dose associated with diagnostic imaging modalities. A fiber-optic coupled point dosimetry system was constructed based on the scintillation properties of a gadolinium oxy-sulfate phosphor coupled to a fused silica optical fiber. The dosimeter system was then evaluated for its performance across the diagnostic energy range and implemented in profiling the dose delivery methods of helical, multi-detector computed tomography (MDCT). The fiber-optic coupled (FOC) dosimeters showed strong sensitivity, reproducibility, and excellent dose linearity in the diagnostic range and provided remote, real-time detection of the dose associated with helical CT. Helical scan dose profiles measured with FOC dosimeters provided information of CT characteristics that traditional computed tomography dosimetry metrics lacks, specifically the isolation of the effects of primary and scatter radiation. The dose profiles display the attenuation and scatter effects of phantom size, length, and position, and quantify the effect of changes in CT scan parameters (pitch and slice thickness) on the absorbed dose to the patient. The evaluation of the computed tomography dose index and its efficiency in predicting the dose delivered from clinically relevant scan lengths has shown that the standard 100 mm long pencil ion chamber synonymous with CT dosimetry fails to sufficiently measure the dose due to scatter radiation for scan lengths longer than 100 mm. The analysis of helical dose profiles show that the CTDI100 metric greatly underestimates dose for large phantom diameters and calls for the use of longer length phantoms for computed tomography dose measurements.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by William Moloney.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Hintenlang, David E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022737:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022737/00001

Material Information

Title: A Fiber-Optic Coupled Point Dosimetry System for the Characterization of Multi-Detector Computed Tomography
Physical Description: 1 online resource (103 p.)
Language: english
Creator: Moloney, William
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Nuclear and Radiological Engineering -- Dissertations, Academic -- UF
Genre: Nuclear Engineering Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The importance of radiation dose and the associated risk from x-ray computed tomography (CT) has been highlighted by the scientific community as recent developments in technology have greatly improved the capabilities of CT scanners. As image quality rapidly improves, it is necessary to quantify the radiation dose delivered from new imaging modalities, and evaluate the means for reducing patient dose to as low a level as possible. Developments in computed tomography technology have thus offered a change in the clinical use of CT, prompting the need to evaluate the radiation risk versus medical benefit and calling into question traditional dosimetry methods that may no longer be applicable to new scanner capabilities. The goal of this work was to develop a prototype point dosimeter for the characterization of the radiation dose associated with diagnostic imaging modalities. A fiber-optic coupled point dosimetry system was constructed based on the scintillation properties of a gadolinium oxy-sulfate phosphor coupled to a fused silica optical fiber. The dosimeter system was then evaluated for its performance across the diagnostic energy range and implemented in profiling the dose delivery methods of helical, multi-detector computed tomography (MDCT). The fiber-optic coupled (FOC) dosimeters showed strong sensitivity, reproducibility, and excellent dose linearity in the diagnostic range and provided remote, real-time detection of the dose associated with helical CT. Helical scan dose profiles measured with FOC dosimeters provided information of CT characteristics that traditional computed tomography dosimetry metrics lacks, specifically the isolation of the effects of primary and scatter radiation. The dose profiles display the attenuation and scatter effects of phantom size, length, and position, and quantify the effect of changes in CT scan parameters (pitch and slice thickness) on the absorbed dose to the patient. The evaluation of the computed tomography dose index and its efficiency in predicting the dose delivered from clinically relevant scan lengths has shown that the standard 100 mm long pencil ion chamber synonymous with CT dosimetry fails to sufficiently measure the dose due to scatter radiation for scan lengths longer than 100 mm. The analysis of helical dose profiles show that the CTDI100 metric greatly underestimates dose for large phantom diameters and calls for the use of longer length phantoms for computed tomography dose measurements.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by William Moloney.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Hintenlang, David E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022737:00001


This item has the following downloads:


Full Text

PAGE 1

1 A FIBER-OPTIC COUPLED POINT DOSIMETRY SYSTEM FOR THE CHARACTERIZATION OF MULTI-DETECTOR COMPUTED TOMOGRAPHY By WILLIAM EDWARD MOLONEY A THESISPRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008William Moloney

PAGE 3

3 Tomy family Mom, Dad, Matt, John, Rob, Liz, Pete and Connor, I love you

PAGE 4

4 ACKNOWLEDGMENTS FirstI would like to thank my advisor, David Hintenlang, Ph.D., for his guidance and patience during my time at the University of Florida. His teaching and tutelage havegiven me the knowledge and direction needed to continue my path in the field of medical physics. He has devoted invaluable time in my research, but more importantly has always held his door open to all his students and it is for his friendship that I will be forever grateful. I would like to thank Dr. Manuel Arreola and Dr. Wesley Bolchfor their time and effort in guiding my graduate career. Their invaluable instruction both in and outside the classroom has provided me with the knowledge and understanding of where I want to be both academically and professionally. I would also liketo thankDr. Alireza Haghighat for the opportunities he has afforded me here at UF. Finally, I need to thank both the faculty and staff inthe Department of Nuclear & Radiological Engineering, especially Diana Dampier and Terri Sparks for making life easier on a daily basis. Withoutthe help and friendship of my fellow students that worked alongside me these past few years, I would not be where I am today. I would like to thank Dr. Kyle Jones, Ryan Fisher, James Winslow, and Dan Hyer for all their support, advice, and most often, their cars. It is without a doubt a reflection of our advisor that we have not only worked together well as colleagues, but have also become such great friends. I would liketoextend no small thank you toDr. Paul J. Angiolilloforintroducing me to the field of medical physics. Another special thank you goes to Dr. David Gandolfo. Their friendship and counsel kept me focused both in and outside academics when I needed it most. Their teaching and advisement helped shape my life more than I can possibly thank themfor and I will never forget it.

PAGE 5

5 Most importantly,I would like to thank my family and friends who mean the absolute world to me, especiallymy parents, Jane andBill, to whom I owe everything. I cannot thank them enoughfor their time, patience, love, and support. They have instilled in theirchildren a love and faith that is immeasurable and we love them more than anything. Finally, I need to thank Elizabeth who keepsme aslevel headed as I can beand who has put up with me somehow these last few months. Ilove her more than words can say.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES...........................................................................................................................9 ABSTRACT...................................................................................................................................12 CHAPTER 1 INTRODUCTION..................................................................................................................14 Computed Tomography Dose.................................................................................................14 Computed Tomography Dose Contributors....................................................................15 Computed Tomography Dose Descriptors......................................................................15 Multiple scan average dose (MSAD).......................................................................15 Computed tomography dose index (CTDI)..............................................................15 Computed tomography dose index (CTDI) phantoms.............................................17 Weighted and volume CTDI....................................................................................17 Dose-length product (DLP)......................................................................................17 Axial Scans.............................................................................................................................18 Helical Scanning.....................................................................................................................18 2 FIBER OPTICS......................................................................................................................21 History of Fiber Optics...........................................................................................................21 Electromagnetic Spectrum......................................................................................................23 Fiber Optic Properties.............................................................................................................26 RefractiveIndex and Total Internal Reflection...............................................................26 Numerical Aperture.........................................................................................................29 Light Collection Efficiency and Transmission................................................................30 Fiber attenuation.......................................................................................................31 Absorption................................................................................................................31 Scattering..................................................................................................................31 Light leakage in the cladding...................................................................................31 Bending losses..........................................................................................................32 3 CONSTRUCTION & CHARACTERIZATION OF A FIBER-OPTIC COUPLED DOSIMETRY SYSTEM........................................................................................................33 Introduction.............................................................................................................................33 System Overview....................................................................................................................33 Optical Fibers..................................................................................................................34 Scintillation Phosphor.....................................................................................................34 Photon Counting Head....................................................................................................37 Photomultiplier Tube.......................................................................................................38

PAGE 7

7 Software...........................................................................................................................39 System Characterization.........................................................................................................40 Energy Dependence.........................................................................................................40 Dose Linearity.................................................................................................................40 Angular Dependence.......................................................................................................41 Table attenuation......................................................................................................42 Radiation incidence..................................................................................................43 Angular dependence in a scattering medium...........................................................46 Conclusions.....................................................................................................................47 4 COMPUTED TOMOGRAPHY DOSE PROFILES..............................................................48 Introduction.............................................................................................................................48 Materials and Methods...........................................................................................................48 Head & Body Phantoms..................................................................................................50 Tube Potential (kVp) & Tube Current-Time Product (mAs)..........................................50 Pitch.................................................................................................................................51 Slice Thickness................................................................................................................51 Results and Discussion...........................................................................................................52 Head Phantom.................................................................................................................52 Body Phantom.................................................................................................................56 Pitch.................................................................................................................................58 Head phantom..........................................................................................................58 Body phantom..........................................................................................................60 Slice Thickness................................................................................................................62 Head phantom..........................................................................................................62 Body phantom..........................................................................................................64 Conclusions.............................................................................................................................67 CTDI Phantom Position..................................................................................................67 Head vs. Body Phantom..................................................................................................68 Pitch.................................................................................................................................69 Slice Thickness................................................................................................................70 5 COMPUTED TOMOGRAPHY DOSE INDEX....................................................................71 Introduction.............................................................................................................................71 CTDI 100 ............................................................................................................................71 Pencil Ion Chamber.........................................................................................................71 Small Volume Ion Chamber............................................................................................72 Material and Methods.............................................................................................................74 Fiber-Optic Coupled Dosimeters.....................................................................................74 CTDI Efficiency..............................................................................................................74 Need for Longer Phantoms..............................................................................................75 Novel method for predicting helical scan dose profiles...........................................76 CTDI L(mm) .................................................................................................................80 Results and Discussion...........................................................................................................81 CTDI 100 Efficiency..........................................................................................................81

PAGE 8

8 CTDI L ( mm) .........................................................................................................................82 CTDI 100 Efficiency Revisited..........................................................................................88 Conclusions.............................................................................................................................90 Novel Method for Predicting Helical Scan Dose Profiles...............................................90 Need for Longer Phantoms..............................................................................................90 6 CONCLUSIONS....................................................................................................................92 Prototypical Fiber-Optic Coupled Point Dosimetry System..................................................92 Helical Computed Tomography Dose Profiles.......................................................................92 Novel Method for Predicting Helical Scan Dose Profiles...............................................94 Need for Longer Phantoms..............................................................................................94 Future Work............................................................................................................................95 Extended Characterization of the Point Dosimetry System............................................95 Verification of Predicted Dose Profiles in Extended Length Phantoms.........................95 Multi-Fiber-Optic Coupled Dosimeter System...............................................................96 Anthropomorphic Phantoms............................................................................................96 Final Thoughts........................................................................................................................97 LIST OF REFERENCES...............................................................................................................98 BIOGRAPHICAL SKETCH.......................................................................................................103

PAGE 9

9 LIST OF TABLES Table page 4-1 Helical scan dose profiles Operating conditions.............................................................49 4-2 Head phantom -Ratios of integrated counts (dose) show the effect of pitch on the helical scan dose profile.....................................................................................................60 4-3 Body phantom -Ratios of integrated counts (dose) show the effect of pitch on the helical scan dose profile.....................................................................................................62 4-4 Head Phantom -Ratios of integrated counts (dose) show the effect of slice thickness on the helical scan dose profile..........................................................................................64 4-5 Body Phantom -Ratios of integrated counts (dose) show the effect of slice thickness on the helical scan dose profile..........................................................................................66 5-1 Helical scan dose profiles Operating conditions.............................................................76 5-2 Calculated percent difference in accumulated dose across a 15 cm scan length for the measured single head phantom profile and the reflected first 7.5 cm of the 300 mm length phantom profile.......................................................................................................80 5-3 CTDI 100 150 ) calculated as the ratio of the accumulated dose measured in the center 100 mm of a standard PMMA phantom divided by the total 150 mm scan length of the phantom........................................................................................................82 5-4. Varia 150 efficiencies calculated from helical scan dose profiles acquired at three pitches for each combination of slice thickness and phantom position utilized.......83 5-5 Accumulated dose ratios measured from helical dose profiles of 450 mm and 150 length PMMA head phantoms...........................................................................................84 5-6 Accumulated dose ratios measured from reflected helical dose profiles of 450 mm and 150 length PMMA head phantoms.............................................................................84 5-7 Accumulated dose ratios measured from helical dose profiles of 450 mm and 150 length PMMA body phantoms...........................................................................................87 5-8 CTDI 100 efficiency (CTDI 100 / CTDI L ) calculated as the ratio of the accumulated dose measured in the center 100 mm of a standard PMMA phantom divided by the total scan length L of the phantom.............................................................................................88 5-9 450 efficiencies calculated from helical scan dose profiles acquired at the three pitches for each combination of slice thickness and phantom position utilized.......88

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Single scan dose profile f(x) x/L ) to produce the cumulative dose profile D(x) where L represents the total helical scan length...........20 2-1 Optical fiber and cladding.................................................................................................22 2-2 The electromagnetic spectrum..........................................................................................24 2-3 Wave-particle duality of light............................................................................................25 2-4 Light refraction in glass....................................................................................................27 2-5 Light refraction through different media...........................................................................28 2-6 Refraction and total internal reflection..............................................................................28 2-7 Total internal reflection......................................................................................................29 2-8 Numerical aperture.............................................................................................................30 2-9 Bending losses...................................................................................................................32 3-1 Diagram of the fiber-optic coupled dosimetry system and component.............................34 3-2 Photo of typical optical fiber dosimeter assembly.............................................................35 3-3 X-ray interactions probability............................................................................................35 3-4 Probability of photoelectric absorption for gadolinium oxy-sulfate as a function of incident photon energy.......................................................................................................36 3-5 Spectral output of gadolinium oxy-sulfate light photons...................................................36 3-6 Photo of photon counting head and housing......................................................................37 3-7 Diagram of photomultiplier tube used in the fiber-optic coupled dosimeter.....................38 3-8 Spectral response of H7467 photon counting head and photomultiplier tube...................39 3-9 Tube potential (energy) dependence of the fiber-optic coupled dosimeter.......................41 3-10 Dose linearity of the fiber-optic coupled dosimeter..........................................................41 3-11 CT tube angular dependence and table attenuation...........................................................42 3-12 Fiber-optic coupled dosimeters angular dependence........................................................44

PAGE 11

11 3-13 Angular dependence of severalfiber optic coupled dosimeters........................................45 3-14 FOCD angular dependence in a scattering medium..........................................................46 4-1 Photograph of the standard PMMA CTDIphantoms.......................................................49 4-2 Comparison of measured data in the five positions of the CTDI head phantom. Center and four peripheral (left and right responses are identical) measurement locations indicated.............................................................................................................53 4-3 Expanded view of the helical dose profile measured in the 12 oclock (top) position of the CTDI head phantom................................................................................................54 4-4 Expanded view of the helical dose profile measured in the center position of the CTDI head phantom...........................................................................................................55 4-5 Comparison of measured data in the five modules of the CTDI body phantom: center and four peripheries.................................................................................................57 4-6 Effect of pitch on measured helical dose profiles in the CTDI head phantom..................59 4-7 Effect of pitch on measured helical dose profiles in the CTDI body phantom..................61 4-8 Effect of slice thickness on measured helical dose profiles in the CTDI head phantom...63 4-9 Effect of slice thickness on measured helical dose profiles in the CTDI body phantom..............................................................................................................................65 5-1 Diagram of the 300 mm length phantom setup..................................................................77 5-2 Helical scan dose profiles in the center position of CTDI head phantoms measured with fiber-optic coupled dosimeters...................................................................................78 5-3 Helical scan dose profiles in the 12 oclock position of CTDI head phantoms measured with fiber-optic coupled dosimeters..................................................................79 5-4 Helical scan dose profiles in the center position of CTDI head phantoms measured with fiber-optic coupled dosimeters...................................................................................86 5-5 Helical scan dose profiles in the 12 oclock (top) position of CTDI head phantoms measured with fiber-optic coupled dosimeters..................................................................87

PAGE 12

12 Abstract of ThesisPresented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science A FIBER-OPTIC COUPLED POINT DOSIMETRY SYSTEM FOR THE CHARACTERIZATION OF MULTI-DETECTOR COMPUTED TOMOGRAPHY By William Edward Moloney August 2008 Chair:David Hintenlang Major: Nuclear Engineering Sciences The importance of radiation dose and the associated risk from x-ray computed tomography (CT) has been highlighted by the scientific community as recent developments in technology have greatly improved the capabilities of CT scanners. As image quality rapidly improves, it is necessary to quantify the radiation dose delivered from new imaging modalities, and evaluate the meansfor reducing patient dose to as low a level as possible. Developments in computed tomography technology have thus offered a change in the clinical use of CT, prompting the need to evaluate the radiation risk versus medical benefit and calling into question traditional dosimetry methods that may no longer be applicable to new scanner capabilities. Thegoal of this work was to develop a prototype point dosimeter for the characterization of the radiation dose associated with diagnostic imaging modalities. Afiber-optic coupled point dosimetry system was constructed based on the scintillationproperties of a gadolinium oxysulfate phosphor coupled to a fused silica optical fiber. The dosimeter system was then evaluated for its performance across the diagnostic energy range and implemented in profiling the dose delivery methods of helical, multi-detector computed tomography(MDCT).

PAGE 13

13 The fiber-optic coupled (FOC) dosimeters showed strong sensitivity, reproducibility, and excellent dose linearity in the diagnostic range and provided remote, real-time detection of the dose associated with helical CT. Helical scan dose profiles measured with FOC dosimeters provided information of CT characteristics that traditional computed tomography dosimetry metrics lacks, specifically the isolation of the effects of primary and scatter radiation. The dose profiles display the attenuation and scatter effects of phantom size, length, and position, and quantify the effect of changes in CT scan parameters (pitch and slice thickness) on the absorbed dose to the patient. The evaluation of the computed tomography dose index and its efficiency in predicting the dose delivered from clinically relevant scan lengths has shown that the standard 100 mm long pencil ion chamber synonymous withCT dosimetry fails to sufficiently measure the dose due to scatter radiation for scan lengths longer than 100 mm. The analysis of helical dose profiles show that the CTDI 100 metric greatly underestimates dosefor large phantom diameters and calls for the use of longer length phantoms for computed tomography dose measurements.

PAGE 14

14 CHAPTER 1 INTRODUCTION The importance of radiation dose and the associated risk from x-ray computed tomography (CT) has been highlighted by the scientific community as recent developments in technology have greatly improved the capabilities of CT scanners. 1-3 With the introduction of helical CT in the early 1990s came the implementation of multi-detector as well as multi-slice systems.4-7 The number of slices acquired per axial rotation has increased greatly from 16-and 64-slice systems to recently developed 256-slice scanners. 8 Developments in computed tomography technology have thus offered a change in the clinical use of CT, prompting the need to evaluate the radiation risk versus medical benefit. According to 2006 data, approximately 62 million CT examinations were performed in hospitals and outpatient imaging facilities in the United States. 9 The dose levels imparted in CT exceed those from conventional radiography and fluoroscopy. Furthermore, its use continues to grow, making CT a significant contributor to the total collective dose delivered to the public from medical procedures involving ionizing radiation. 10-11 Computed Tomography Dose Mettler etal. state that the increasing use of computed tomography in clinical practices coupled with its relatively high radiation doses compared to general radiography makes CT a significant contributor to patient population dose from medical x-rays. 10-11 It is therefore necessary to establish an accurate and easily reproducible method for CT dosimetry. Fundamental definitions of CT dose parameters require review and possibly reinterpretation with advances in CT technology. Traditional CT parameters have not been used consistently (or are even out of date), while some may be more relevant than others with respect to patient risk or new scanner capabilities.

PAGE 15

15 Computed TomographyDose ContributorsFour aspects of CT image acquisition are unique in comparison to x-ray projection imaging. First, the volume of tissue being irradiated by the primary beam is much smaller. Second, the volume of tissue being irradiated is exposed from almost all angles, and thus the dose is more evenly distributed to the tissues in the beam. Third, CT requires a high signal to noise ratio to achieve higher contrast resolution, and therefore the higher techniques (kV and mAs) lead to higher dose to the slice volume. Finally, scatter radiation is a significant contributor to CT radiation dose. Compton scatteringis the principle interaction mechanism in CT and the scatter dose can often be higher than that of the primary beam. The acquisition of a CT slice also delivers a considerable amount of dose from scatter to adjacent tissues outside the collimated primarybeam.12 Computed TomographyDose DescriptorsQuality assurance programs and optimization procedures in diagnostic radiology require the assessment of patient doses. Commonly adopted dose descriptors based on phantom measurements that are easy to measureand define are described below. They include the MSAD and the CTDI. Multiple scan average dose (MSAD) The multiple scan average dose (MSAD) is the standard for CT radiation dosimetry. It is defined by the International Atomic Energy Agency (IAEA) as theaverage dose, at a particular depth from the surface, resulting from a large series of CT slices. 12-13 It is the dose to tissue including the dose that is attributable to scattered radiation from all adjacent slices. Computed tomography dose index (CTDI) An estimate of the MSAD can be accomplished with a measurement of the CTDI (Computed Tomography Dose Index). 14 CTDI FDA is defined by the U.S. Food and Drug

PAGE 16

16 Administration as the radiation dose to any point in the patient including the scatter radiation contribution from 7 CT slices in both directions, for a total of 14 slices. 15 One method used to measure the CTDI FDA is with an array of thermoluminescent dosimeters (TLDs) placed in holes along 14-slice thickness increments in a specified CTDI phantom. A single CT sliceis acquired at the center of the array, and the CTDIFDA is determined by summing the individual TLD dose measurements. Another CTDI FDA measurement method requires the use of a thin pencil ionization chamber. 16-17 Whether it is an ion chamber or TLD array, the length of the detector must be long enough to average the variations produced by the scan interval, but not so long that the average is distorted by the dose fall off at the edge of the scan length. As a result, CTDI FDA can greatly underestimate the MSAD for small slice thicknesses, because a significant amount of radiation is scattered beyond seven slice thicknesses and CTDI FDA defines no standardization of slice width in determining dose measurements. A much better CT Dose Index has been defined by the IEC as the exposure measured over a 100 mm length for all slice thicknesses. 18 This CTDI 100 uses a 10 cm long, 3 cc pencil ion chamber and a single CT sliceis producedat the center of the chamber in the standardized CTDI phantomsdescribed below. The active area of the pencil chamber provides anexposure reading that represents an average exposure (air kerma) over the chamber length (100 mm). CTDI 100 (rad or Gy) = (mm) ) ( reading exposure (mm) 100 (rad/R) T N R f C (1-1) where R is the exposurereading,T is the slice thickness, N is the number of slices, C is the temperature and pressure corrected chamber calibration factor, and f is the appropriate exposureto-dose conversion factor.

PAGE 17

17 Computed tomography dose index (CTDI)phantomsSpecified CTDI phantoms are composed of polymethylmethacrylate (PMMA) and are 15 cm in length, with diameters of 16 and 32 cm corresponding to the head and body phantoms, respectively. 2-3,19 There are five drilled holes parallel to the z axis (couch direction) to allowfor the placement of dosimeters for the measurement of the central and peripheral (top, left, right, and bottom,with respect to the couch and at 1 cm from the surface) CTDI. Weighted and volume CTDI The weighted CTDI is defined as CTDI(W) = 3 1 CTDI (C) + 3 2 CTDI (P) (1-2) where CTDI (C) is the central CTDI and CTDI (P) is the average of all four peripherals. 20-21 The volume CTDI (CTDI vol ) is then used to represent the dose for a specific scan protocol, usually involving a series of scans and taking into account protocol-specific information such as pitch. It is defined as CTDI (vol) = I T N CTDI (W) = pitch 1 CTDI (W) (1-3) where I is the table increment (mm) per axial scan. Dose-length product (DLP) CTDI vol only estimates the average dosefor a 100 mm scan lengtheven though the actual volume-averaged dose will increase with scan length. A better representation of the overall energy delivered by a given protocol is to integrate the absorbed dose along the scan length. This metric is known as the Dose-Length Product (DLP): DLP (mGy cm) = CTDI vol (mGy) scan length (cm) (1-4)

PAGE 18

18 The DLP thus reflects the total energy absorbed due to a complete scan acquisition and offers a potential for quantifying biological effect. Axial Scans For a narrow, fan-shaped beam of x rays rotating around a CT phantom in the (x,y) plane producing a single slice image (or for multiple slices through consecutive rotations with couch movement in the z direction between scans), the CTDI may be defined by the formula CTDI = T 1 D (z) dz (1-5) where D(z) is the single, axial scan dose profile (sum of the primary radiation within the imaged volume and the scattered radiation within and outside the imaged volume) along a line parallel to the z axis of the scanner for nominal slick thickness T 22 For multiple slices through consecutive rotations with couch movement in the z direction between scans, the formula requires division by the number of slices N in a single scan. In clinical applications, the integration in Eq. (1-5) is carried out for either a total thickness of 14 slices (CTDI FDA ) or 100 mm (CTDI 100 ). 23-24 While the CTDI 100 remains an accurate prediction of MSAD for axial CT, it begins to fail for multiple scan dose profiles of pitch measuring the integral of the single-slice dose profile using a 10 cm long ion chamber to predict the MSAD can greatly underestimate dose as radiation beam widths for multi-slice scanners get wider. If the ion chamber is not long enough to integrate the total areas under the single slice dose profile (Eq.3-4), the scatter tails of the beam profile are omitted, thus underestimating the dose significantly. 25 Helical Scanning Metrics,introduced by Robert Dixon based on the use of convolution mathematics to evaluate CT scan dose profiles provide a method of dosimetry where the 10 cm ion chamber

PAGE 19

19 fails. 26 In helical CT, continuous translation of the single-scan dose profile results in a continuum of contributions to the cumulative dose profile. To visualize this, consider a helical scan in which the table and phantom are translated along the longitudinal axis with constant velocity Let f(x) be the single slice axial dose profile that travels a length L of the phantom (from L/ 2 to L /2). The cumulative dose profile can thus be obtained by integrating across the entire scan length, D(x) = 1 2/ 2/L L f(x) dx(1-6) where is the time for one rotation of the scanner. Figure 1-1 displays the single slice dose profile f(x) It can be seen that the maximum dose, D(x=0) occurs in the center of the scan length. The right side of the figure displays the convolution of the single slice dose profile f(x) x/L ), which has unit height and width L The cumulative dose profile D(x) builds up as the single slice profile slides into the box (L /2, L /2) and the product integrated. The dose in the central region flattens out and reaches an equilibrium value when L is large enough to encompass the scatter tails of f(x) 26-27 The equilibrium dose value is then given by D eq (0)= 1 f(x) dx(1-7) Physically measured dose profiles will have an accumulated dose D(x) that is nonuniform. Instead it is useful to obtain the line integral of the dose along the axis. This metric is the dose line integral (DLI) and is analogous to the dose-length product discussed earlier. 26 DLI = D(x) dx = 1 f(x) dx = L Deq (0) (1-8) The infinite integration limits represent the true dose-length product and includethe scatter tails distributed beyond the interval (L /2, L /2).

PAGE 20

20 Figure 1-1. Single scan dose profile f(x) x/L ) to produce the cumulative dose profile D(x) where L represents the total helical scan length The cumulative dose profile and the dose line integral are thus generated from the single slice dose profiles using aconvolution across the scan lengthand scan lengths utilized by most clinical exams are long enough such that the dose equilibrium in the central region should be reached. 26 Thus, the equilibrium dose can be measured with a single dose profile if the chamber is long enough to encompass all the scatter tail radiation resulting from the scan. The use of a small volume ion chamber to directly measure the cumulative dose D(x) at any point by scanning a length of phantom long enough to produce dose equilibrium in the center has the same result as making the chamber longer and is indeed more accurate for wide beam profiles. Even if the scan length L is not long enough to produce equilibrium at the center, such a small chamber will give a good measurement of the maximum dose, since the dose distribution is relatively flat in the center, and a 10 cm chamber will merely give the average dose over the central 10 cm of the scan length. Theintroduction of a small ion chamber to measure the flat, center of the helical dose profile allows for the determination of important dose profile characteristics and thus leads to a more accurate prediction of MSAD for cone-beam, helical computed tomography.

PAGE 21

21 CHAPTER 2 FIBER OPTICS Though they have risen into a booming industry inthe field of communications technology, fiber optics evolved from devices designed to guide light for illumination. They were first used to look inside the human body, and are still used todayas imaging tools in endoscopy and colonoscopy. Historically,optical fibers have also offered a unique capability of radiation monitoring. The development of a fiber-optic coupled dosimetry system will show further use of this rising technology in medical applications, specifically in the area of radiation detection and dose calculation. This point detector will provide remote, real-time dose measurements and allow for the direct recording of the radiation characteristics of modern x-ray technologies. History of Fiber Optics In 1880, engineer William Wheeler patented a way to pipe light through buildings. He wanted to distribute light from a light source known as an electric arc to distant rooms within a building through a set of pipes coated with a reflective layer, where diffusers at the end of each pipe would spread the light out. At the time, air was a much clearer medium than any known solid, and though logical, his idea never caught on. However, by the early 1900s, a scheme for bending light through a bent glass rod was created to illuminate the inside of the mouth for dentistry. 28 Through a phenomenon known as total internal reflection, light can be confined inside glass or another transparent material. The light is sent through the material in such a way that it strikes the surface exposed to air at a glancing angle. The light is then reflected back into the solid. A fine glass fiber is actually a very thin, flexible rod, so it can guide light in much the same way.

PAGE 22

22 The use of fiber optics for medical imaging purposes took its first steps in the 1920s, as glass fibers were assembled into bundles. The first image-transmitting bundles were not consistently effective due to the use of bare fibers. As the fibers touched and scratched, light was able to leak at the surface whenexposed to air. Total internal reflection occurswhen light traveling in one medium tries to enter another medium with a lower refractive index. Air has a much lower refractive index than glass, but the difference need not be large. Coating the glass fiber with such a material can allow for total internal reflection, while still protecting the surface of the fiber from scratches, fingerprints, and light leakage into other glass fibers (Figure 2-1). 28 Figure 2-1. Optical fiber and cladding In the 1950s, scientists, engineers, and students alike discovered this more practical use of fiber bundles and went as far as coating the glass rods with oils, margarine, beeswax, and finally different plastics. In December 1956, a University of Michigan undergraduate student by the name of LarryCurtiss slipped a rod of glass with a high refractive index into a lower indexed tube of glass, making the first glass-clad fiber. Technology has improved since then, but glassclad fibers remain in common use, and fiber bundles have been used extensively in imaging and illumination. 28

PAGE 23

23 The early use of optical fibers was limited by signal loss as the original endoscopes lost half of the light they carried after 3 meters. With the same fibers, only 0.1% of the light remained after 30 meters. While this can be acceptable for use over short distances, long distance applications such as in communications would be deemed impossible. It was in the late 1960s that two engineers at Standard Telecommunications Laboratories in England,Charles K. Kao and GeorgeHockham, discovered that most of the loss in fiber glass was due to impurities and the development of highly purified glass would prove them right. Todays best fibers allow for 10% remaining light after a distance of 50 kilometers. 28 Electromagnetic Spectrum Understanding the fundamental properties of fiber optics requires a basic review of the principles that guide lights interaction with matter in the field of optics. Optics is the branch of physics that describes the behavior and properties of light and the interaction of light with matter. Light makes up only a small part of the spectrum of electromagnetic radiation and the waveparticle duality of modern physics allows us to consider light to be either electromagnetic waves or photons that travelat the speed of light, c (2.998 108 m/s). The difference between radiations in the electromagnetic spectrum can be measured in several ways: the length of the wave, the energy of a photon, or as the oscillation frequency of an electromagnetic field (Figure 2-2). Viewed as an electromagnetic wave, light is composed of electric and magnetic fields, which are perpendicular to each other and to the direction in which the light travels as shown in Figure 2-3a. The amplitude of each field varies sinusoidally. The wavelength, ,is defined as the distance that light travels during one complete cycle (rising from zero to a positive peak, going back through zero to a negative peak, then returning to zero) and is measured in meters, micrometers ( m or 10 -6 m), or nanometers (nm or 10 -9 m). The frequency, (measured in

PAGE 24

24 hertz), is described as the number of waves or cycles per second (cps). Wavelength decreases as frequency increases, and waves can be characterizedby either. Figure 2-2. The electromagnetic spectrum. A photon is a quantum of electromagnetic energy or wave packet, a series of waves that build quickly to a peak amplitude, and then fade back to nothing (Figure 2-3b). The pulse or wave packet has the same characteristics as the electromagneticwaves described above; however, the wavelength and frequency are not as well defined as in a continuous sine wave. The uncertaintyprinciple tells us that the shorter the pulse, the larger the uncertainty in wavelength. The amount of energy carried by a single photon depends on the oscillation frequency. The faster the wave oscillates, the higher the energy. Thus, each photon has a unit of energy set by the wavelength or frequency. A continuous wave is then a series of photons emitted one after the other, and the total energy is the number of photons times that photon energy. In wave terms, that is proportional to the wave amplitude squared.

PAGE 25

25 Figure 2-3. Wave-particle duality of light. A) Wave consisting of electric and magnetic fields. B) A photon, quantum of energy, or wave packet. Referring back to Figure 2-2, all the measurement units on the spectrum chart are different ways to measure the same thing, and there are simple conversion methods between them. Wavelength is inversely proportional to frequency, wavelength = frequency c or (m) = ) ( ) / ( Hz s m c (2-1) Photon energy can be measured from Plancks law, which states (2-2) where h is Plancks constant (6.626 10 -34 Js, or 4.136 10 -15 eVs). But since our interest lies in the part of the spectrum measured in wavelength, a more useful formula is E ( eV ) = ) ( 2399 .1m (2-3) which gives energy in electron volts (the energy that an electron gains in moving through a 1 volt electric field) when wavelength is measured in micrometers ( m). Looking again at Figure 2-2, the interest of this paper is in the region labeled The Optical Spectrum, where optical fibers function. This region includes the light visible to the human eye

PAGE 26

26 (400 to 700 nm) and nearby parts of the infrared or our interest, the ultraviolet, where fused silica fibers can transmit light over short distances. Fiber Optic Properties Refractive Index and Total Internal Reflection The characterization and understanding of the path that light travels has been the precursor to the development of fiber optic applications. There is a noticeable difference of speed as light travels through varying transparent materials. The refractive index, n is the speed of light in a vacuum, c divided by the speed of light in a material: n = material vacuumc c (2-4) The change in speed of light when traveling from one materialinto another is an effect known as refraction. Figure 2-4 depicts what happens to the peaks of light waves as they enter glass from air. The waves in air continue at the same speed until they reach the surface of the glass, where they slow down. As thewaves hit the glass at an angle, some light enters the glass while the rest remain in the air. The frequency of the wave does not change as the waves slow down in glass, but the distance traveled between peaks is shortened. 28 This slowing process bendsthe path of light. The amount of bend that the light experiences depends on the refractive indexes of the two materials and the angle of incidence at the surface. Snells law states: n i sin I = n r sin R (2-5) where n i and n r are the refractive indexes of the initial medium and the medium into which the light is refracted, while I and R are the angles of incidence and refraction of the transmitted light relative to the normal, as seen in Figure 2-4.

PAGE 27

27 Figure 2-4. Light refraction in glass The above description explains the standard example of light traveling from air into glass. The opposite, light emerging into air from glass, is shown in Figure 2-5. The process of refraction is reversed, and if the front and rear surfaces of a glass window are flat, the net refraction is zero. If we look at the refraction of a lens, however, light emerges at a different angle than when it entered the glass. When light in a medium with a high refractive index (glass) interfaces with a medium of low refractive index (air), the light is bent farther away from the normal; i.e. according to Snells law, the angle of refraction is greater than the angle of incidence ( R > I, or 2 > 1 ). This is not a problem for small angles of incidence, but when the angle of incidence becomes too large, refraction cannot occur. 28 If the angle of incidence exceeds a critical angle, where the sine of the angle of refraction would be equal to 1.0, light cannot get out of the glass. Instead total internal reflection occurs as shown in Figure 2-6. The critical angle above which total internal reflection takes place can be derived from Snells law, critcal = arcsin ( i r n n ) (2-6)

PAGE 28

28 Figure 2-5. Light refraction through different media. A) Window. B) Lens. Figure 2-6. Refraction and total internal reflection

PAGE 29

29 Numerical Aperture Optical fibers are made of glass which can be inherently strong when not cracked and extremely flexible when thin. The comparison is often made to human hair, and though fibers are much stiffer, when well made, they can be much smoother than hair. A small light-guiding core, which can be as thin as a few microns, is surrounded by a glass cladding. A plastic coating covers the entire fiber, protecting its surface from scratches and microcracks that can cause signal loss or the fiber itself to break. The refractive index of the core is higher than that of the cladding, so light striking the interface at a glancing angle is confined in the core by total internal reflection, Figure 2-6. The difference, however, need not be large. For a 1% difference (i.e. n r /ni = 0.99), the critical angle, critical ,measured from the normal is approximately 82. Figure 2-7 shows us that any light striking the cladding at an angle of 8 or less will be confined in the core. This upper limit is confinement of the fiber. 28 Figure 2-7. Total internal reflection Another wayto characterize the light guiding capabilities of an optic fiber is by looking at the angle over which light rays entering the fiber will be guided successfully along its core, Figure 2-8. This acceptance angle differs from the confinement angle described above because it

PAGE 30

30 is instead measured in air. The acceptance angle is measured as the numerical aperture (NA), which for light entering a fiber from air is approximately NA = ) ( 2 2 I O n n (2-7) where n 0 is the refractive index of the core and n 1 is the refractive index of the cladding. Again using the example of a 1% difference between core ( n 0 = 1.5) and cladding ( n 1 = 1.485), NA = 0.21. As seen in Figure 2.8, refraction bends a light ray entering the fiber so that it is at a smaller angle to the fiber axis than angle (12), is larger than the confinement angle (8), with the difference being the refractive index of the core. Thus another way to define the numerical aperture of a fiber is NA n 0 confinement (2-8) Light Collection Efficiency and Transmission Collecting light efficiently is an obvious desire when using optical fibers for detection purposes. Coupling light effectively into the fiber, whether from an external source or from one fiber to the another, requires both focusing external light onto the core and aligning within the fibers acceptance angle. Other inherent factors in optical fibers that contribute to some signal loss include absorption, scattering, and leakage of light from the fiber core. Figure 2-8. Numerical aperture

PAGE 31

31 Fiber attenuation The attenuation of an optical fiber measures the amount of light lost between input and output. Total attenuation is the sum of all losses. It is dominated by imperfect light coupling into the fiber and absorption and scattering within the fiber. Absorption and scattering are both cumulative over the length of a fiber, increasing with distance. Coupling losses, however, occur only at the ends of the fiber. Forlong fibers, absorption and scattering losses dominate. Conversely, they may be much smaller for shorter fibers, where coupling losses dominate. Absorption Although optical fibers are made of pure glass, they still absorb a tiny fraction of the light passing through them. How much depends on the wavelength of the light and any impurities in the glass. While impurities can cause strong absorption, even pure silica has some absorption while transmitting most of the light that enters it; however, absorption is uniform. The same amount of material absorbs the same amount of light at the same wavelength. Absorption is also cumulative. A material absorbs the same fraction of light for each unit length. Scattering Scattering reflects light in other directions, so it escapes from the fiber core and is lost from the signal. The atoms in the glass cause Rayleigh scattering of the light waves. Scattering depends not on the specific type of material, but on the size of the particles relative to the wavelength of light. The closer the wavelength is to the particles size, the more scattering. Like absorption, it is inherent in all fiber materials, but is generally small, increasing at shorter wavelengths. Scattering is also both uniform and cumulative as in fiber absorption. Light leakage in the cladding Light leakage also occurs when light escapes from the fiber core into the cladding at an angle greater than the confinement angle discussed earlier. To prevent this, fibers are coated

PAGE 32

32 with a plastic that has a higher refractive index higher than that of the cladding, preventing total internal reflection. Figure 2-9. Bending losses Bending losses Significant losses can also arise if a fiber is bent so sharply that light strikes the cladding interface at a largeenough angle that the light can leak out. When the fiber is straight, light falls within the confinement angle. Bending the fiber changes the angle at which light hits the corecladding boundary, as seen in Figure 2-9. If the bend is sharp enough, light hits the boundary at an angle outside the confinement angle, and is refracted into the cladding where it can leak out. There are two types of bending loss. Macrobends are single bends, visible to the naked eye, while microbends are tiny kinks or ripples that can form along the length of the fiber.

PAGE 33

33 CHAPTER 3 CONSTRUCTION& CHARACTERIZATION OF A FIBER-OPTIC COUPLED DOSIMETRY SYSTEM Introduction Optical fibers have offered the unique capability of remotely monitoring radiation in difficult to access and/or hazardous locations. 29 Optical measurement methods for characterizing absorbed radiation dose have included optical attenuation (density), luminescence, optically stimulated luminescence (OSL), thermoluminescence and scattering. 29-32 The following sections describe the development of a prototypicalpoint dosimetry systemthat utilizesfiber-optic coupled (FOC) dosimeters to measure fundamental parameters associated with radiation dosimetry. This point detector approach provides remote, real-time dosemeasurements and allows direct recording of the radiation characteristics of x-ray and computed tomography (CT) imaging modalities. System Overview FOC dosimeters based on the sensitive elements of acoupled scintillation phosphor are characterized for their performance across the diagnostic energy range based on energy dependence, dose linearity, and angular response. This new type of dosimetry system (Figure 31) is based on the detection and absorption of x-rays in a scintillation phosphor material that is coupled to an optical fiber. The phosphor provides a conversion of x-ray photons to visible light, which then travels the length of the fiber to a photomultiplier tube (PMT). The PMT converts the incident light photons to an output voltage signal. Theoutput is proportional to the number of light photons incident at the photomultiplier tube interface. This output is then relayed in terms of (photon) counts to a PC.

PAGE 34

34 Figure 3-1. Diagram of the fiber-optic coupled dosimetry system and component (highlighted portion displays the fiber assembly) Optical Fibers The FOC dosimeter utilizes 400 m diameter, pure silica core fibers 33 purchased from Ocean Optics, Inc. i Surrounding the core is a doped-fluorine silica cladding. 34 A diagram of the fiber assembly is provided in the zoomed in portion of Figure 3-1. A jacketing of standard black heat shrink tubing is then applied over the core andcladding to both strengthen the fiber and reduce the affect of stray light from entering or exiting the fiber assembly. The fibers are interfaced with the PMT housing by standard SMA 905 fiber optic connectors. Figure 3-2 is a digital photo of a typical dosimeter assembly used in the preceding measurements. Scintillation Phosphor Coupled to the end of the optical fiber is a scintillation phosphor material which provides a conversion of x-ray photons to visible light. The phosphor material absorbs (detects) incident xi Ocean Optics Inc., 830 Douglas Avenue, Dunedin, FL, 32698 USA.

PAGE 35

35 rays. When the x-rays are absorbed in the scintillator and deposit their energy, a certain fraction of the energy is converted and visible or ultra-violet (UV) light is emitted. Figure 3-2. Photo of typical optical fiber dosimeter assembly Figure 3-3. X-ray interactions probability (photon energy vs. atomic number) Figure 3-3 shows that although the probability of photoelectric absorption decreases with increasing x-ray energy, it increases with increasing atomic number of the absorption material. The probability of photoelectric absorption per unit mass is approximately proportional to ( Z 3 /E 3 ), where Z is the effective atomic number of the absorption material and E is the energy of the incident photon. 35 Thus, for high Z materials the photoelectric effect is the dominant interaction for x-rays in the diagnostic energy range (Figure 3-3). Therefore, gadoliniumoxy-

PAGE 36

36 sulfate (Gd 2 O 2 S) was the scintillation phosphor of choice due to its higheffective atomic number ( Z eff = 50.3), its common use in diagnostic radiology film,and its excellent x-ray absorption efficiency in the diagnostic energy range (Figure 3-4). 36-38 Figure 3-4 Probability of photoelectric absorption forgadolinium oxy-sulfate as a function of incident photon energy Figure 3-5. Spectral output of gadolinium oxy-sulfatelight photons (peak at 550 nm) [Reprinted with permission 2006 Carestream Health, Inc.]

PAGE 37

37 Gadolinium oxy-sulfates ability to converta relatively small number of x-ray photons to a large number of light photons is due predominantly to x-ray absorption via the photoelectric effect in the high Z components of the phosphor dominated by gadolinium ( Z = 64). X-ray absorption by the phosphor leads to the emission of visible light, primarily at wavelengths of 550 nm (Figure 3-5). 36-37 Theselight photonsthen traverse the length of the fiber and are collected by a HAMAMATSU photon counting head. Figure 3-6. Photoof photon counting head and housingPhoton Counting Head The H7467 series Hamamastsu ii photon counting head contains a photomultiplier tube, high voltage power supply circuit, photon counting circuit, 20-bit counter and microprocessor all within a common metal housing. 39 An RS-232 interface allows data transfer and measurement integration time to be controlled through a PC and an optical fiber adapter allows connection with a fiber assembly for light input. Figure 3-6 displays the H7467 photon counting head and ii Hamamatsu Photonics K. K., 325-6,Sunayama-cho, Naka-ku, Hamamatsu City, Shizuoka Pref.,430-8587, Japan.

PAGE 38

38 associated components. The 5-volt powersupply and photon counting head are housed in an aluminum casing for improved durability and shielding against ambient light and scatter radiation. Photomultiplier Tube The photomultiplier tube (PMT) allows the conversion of light photons to a corresponding electrical signal. The scintillation phosphor absorbs incident X-rays and converts them to visible light photons. These photons then traverse the length of the optical fibers as described in Chapter 2, where they enter the photomultiplier tube. Figure 3-7 is a diagram of the photomultiplier tube used in the fiber-optic coupled dosimeter. 39-40 Figure 3-7. Diagram of photomultiplier tube used in the fiber-optic coupled dosimeter Light enters the PMT through a borosilicate glass window, which allows radiation transmission from the infrared to approximately 300 nm. Deposited on the inner surface of this entrance window is a bialkali photocathode. This photosensitive area absorbs the transmitted light and emits photoelectrons into the vacuum created bythe glass housing. The photoelectrons emitted by the photocathode are then focused by electrodes to the dynodes, where electron multiplication occurs. The multiplied electrons are collected by the anode where an output

PAGE 39

39 signal is produced. This output signal is proportional to the amount of light incident on the photocathode. The photon countingheadthen relays the output (number of light photons collected) as counts where it is read out at a PC. The manufacturer provided specifications for the Hamamatsu H7467 PMT shows that the bialkali photocathode has a highly sensitive spectral response in the range of 300-650 nm, i.e. the visible range of the electromagnetic spectrum (Figure 3-8). 39 This increased sensitivity in the visible spectrum makes Gd 2 O 2 S a good match for photocathode coupling due to its primary light emission at wavelengths of 550 nm (Figure 3-5). 36-39 Figure 3-8. Spectral response of H7467 photon countinghead and photomultiplier tube [Reprinted with permission 1999 Hamamatsu K.K.] Software The fiber optic dosimeter system is run using the MS-DOS program provided by the manufacturer of the Hamamatsu H7467 Photon Counting Head, with small custom modifications. The program allows the user to choose and set an integration time for the counter and define a file path for the raw data (counts). A macro-enabled MS Excel program allows

PAGE 40

40 analysis of raw data for experimental purposes. Measured counts are typically plotted versus time, and an integrated sum is provided. System Characterization System characterization tests across the diagnostic energy rangewere performedto evaluate individual fibersenergy dependence, dose linearity, and angular response. All tests were done at Shands Orthopedic Institute at the University of Florida. The clinical x-ray tube was a standard table-top x-ray unitwith a half value layer (HVL) of 3.7 mm aluminum equivalent. The clinical CT unit is a Siemens Somatom Sensation 16 slice scanner. Energy Dependence The energy dependence of the FOC dosimeter was tested by simultaneously exposing the active area of the fiber and a 15 cc Keithley Model 96035B dual entrance window pancake ionization chamber with a Keithley Model 35050A electrometer. The tube current-time product was held constant at 100 mAs for all irradiations, while tube potential (kVp) was incremented within the range of 50 to 120 kVp. Three exposures were performed at each selected tube potential. The average of the three exposures wasthen calculated and plotted versus corresponding kVp values recorded fromthe Kiethley electrometer. Variations in kVp were minimal from exposure to exposure, withstandard deviations not exceeding 0.12. Figure 3-9 is a plot of the fiber optic systems energy response across the diagnostic range with vertical error bars representing plus-and-minus one standard deviation. Dose Linearity The dose linearityof the FOC dosimeter was tested by successively increasing the tubecurrent time product while maintaining a constant tube potential at 80 kVp. Again, the average of three exposures was taken for each tube current setting. FOC dosimeter responseis plotted in

PAGE 41

41 Figure 3-10 with error bars representing plus-and-minus one standard deviation. A linear fit shows that the fiber response islinear across therange of 20 to 200 mAs. Figure 3-9. Tube potential (energy) dependence of the fiber-optic coupled dosimeter Figure 3-10. Dose linearityof the fiber-optic coupled dosimeterAngular Dependence Qualifying the fibers response to the incidentangle of radiation was less straightforward then the above characterization tests. The energy dependence and dose linearity evaluations were performed using a standard table-top x-ray unit. However, fiber optic dosimeter angular

PAGE 42

42 dependence was assessed using the Siemens Somatom Sensation 16 slice CT scanner at all angles and involving different experimental parameters. In order to isolate any angular dependence to the response of the fiber alone, several experimental factors had to be characterized. First the CTs beam delivery method was analyzed. A calibrated 10 cm pencil ion chamber was placed in isocenter off the edge of the CT table and exposure measurements were taken free-in-air every 10 degrees. Figure 3-11plots the normalized exposure for the free-in-air ion chambermeasurements. The data series labeled free-in-air shows that the intensity of the CTs x-ray delivery is constant regardless of tube angle. Figure 3-11. CT tube angular dependence and table attenuation. Measurements were made free-in-air with a 10 cm long pencil ion chamber placed both off and above the CT table. Table attenuation Figure 3-11 also shows the effect of attenuation from the CT table. A series of exposure measurements was performed and plotted for a 360 tube delivery; however, the ionchamber

PAGE 43

43 was located at isocenter approximately 8 cm above the surface of the CT table (the height of the center position of a typical CTDI head phantom). The exposure measurements performed above the table were then normalized to the free-in-air results measured off the table. Figure 3-11 shows significant table attenuation of approximately 15% directly below the table and up to 33% just below the sides of the table. This increase in attenuation is due to the structure of the sides of the table. Note that in Figure 3-11, 270 corresponds to the 12 oclock position, and thus 90 refers to when the tube position is located directly under the CT table. Radiation incidence The ion chamber measurements proved that the CT tube delivery is uniform about a 360 rotation and that table attenuation of the primary beam must be taken into account. Thus, further tests on the response of the fiber optic system to tube angle position would confirm an angular dependence unique to the fiber alone. As discussed in Chapter 2, incident light must fall within an acceptance angle in order to be transmitted down the length of the fiber by total internal reflection. The scintillation material coupled to the fibers end should allow for the majority of converted light photons to fall within this acceptance window and be incident on the fiber. Therefore, the directionoftheradiation incident on the cross-sectional area of the phosphor should determine the amount of light produced. In other words, the larger the area struck by incident radiation the more x-rays are absorbed. The larger the cross-section available for photoelectric absorption, the more light photons can be produced, thus strongly influencingthe number of counts collected at the PMT. In MDCT scanners, the arc of detectors and x-ray tube rotate together around the CT table. The table moves or translates along the z-axis, perpendicular to the x-y (axial) CT plane. The

PAGE 44

44 fibers angular dependence was tested free-in-air with regards to two orientations of beam incidence relative to the dosimeter: axial and perpendicular-to-axial. In theformeror head-on approach, the fiber is oriented to face the 270 tube position (12 oclock), parallel to the axial plane. Whenthe x-ray tube is at this orientation, radiation will be incident on the largest available area of the scintillation phosphor (i.e. parallel to the fiber). Figure 3-12shows normalized counts versus x-ray angle of incidence. A major reduction in counts (approximately 46%) is seen when the tube is at the 90 position (i.e. directly below the table and 180 from the surface of the fiber). A less severe but still apparent degradation in counts can be seen at the 0 and 180 tube angles (perpendicular to the fibers surface), with the maximum count intensity occurring at a head-on incidence of the fiber (i.e. the 270 tube position). Note in Figure 3-12, the initial tube position is at 0or 360 (3 oclock). Figure 3-12. Fiber-optic coupled dosimeters angular dependence. Measurements were made free-in-air with a FOC dosimeter oriented perpendicular and parallel to the axial ( x-y ) CT plane. The remaining tests of dosimeter performance were run with the fiber positioned along the zaxis, perpendicularto the axial CT plane. As seen in Figure 3-12, the fiber has an

PAGE 45

45 approximately sinusoidal response, unique when compared to the axial or head-on approach. In the axial orientation, the maximum count intensity occurred when the x-ray tube was positioned above the table (180-360 position) and the incident radiation was directed parallel to the fibers active area. The perpendicular-to-axial response in Figure 3-12, however, shows a minimal fiber response in this range. In fact, when compared with other coupled fibers that were constructed, the perpendicular-to-axial responses all show a similar sinusoidal response with minima at different phases (Figure 3-13). Figure 3-13. Angular dependence of several fiber optic coupled dosimeters. Measurements were made free-in-air with the dosimeters oriented perpendicular to the axial ( x-y ) CT plane. This phase difference when comparing fibers is due to the coupling of the phosphor material to the silica fiber endduring the construction process. Thejacketingofheat-shrink tubing around the fiber core and cladding not only strengthens the fiber and protects it from stray radiation, but helps hold the coupled scintillation phosphor in place. During the jacketing

PAGE 46

46 process, the coupled phosphors orientation may bedisplaced. If this occurs, the dosimeters sensitivity may change depending on the angle of incident radiation. Angular dependence in a scattering medium FOC dosimeters placed free-in-airshow an angular dependence contingent upon the orientation of scintillation phosphor coupling to the fiber end of the dosimeter. However, dose measurements are not typically performed free-in-air, but within some type of tissue-equivalent medium. Taking this into account, the trans-axial angular dependence of the fiber was re-tested. Figure 3-14. FOCD angular dependence in a scattering medium The fiber was placed in the center of a cylinder of tissue equivalent material (radius = 2.5 3 ). Figure 3-14shows the angular dependence of the fiber when placed in sucha small sized scattering mediumwhen compared to the dimensions of typical CT dosimetry phantoms. The angular variationof the fiber response is significantly reduced. When the cylinder is mounted off the tables edge, Figure 3-14displays a full 360 characterization of fiber

PAGE 47

47 angular dependence. When placed in isocenter above the table, the significance of CT table attenuation (0 180 on the x -axis) is again apparent (up to 30%). Conclusions Fiber optic dosimeters coupled with scintillation phosphor material have a strong sensitivity in the low kV energy range and show strong promise for accurate dose measurements in diagnostic radiology. Their energy dependence, specifically a sensitivity to lower energies in the diagnostic range, must be taken into account and dose measurements must be calibrated individually for the energy technique chosen. An inherent angular dependence seen in the fibers coupled active area can be significantly reduced with perpendicular-to-axial fiber positioning and is negligible when the dose measurements are performed within a scattering medium. The fiberoptic coupled dosimeters demonstrate high sensitivity, reproducibility, excellent dose linearity, and combined with their small physical size permit accurate point-dose measurements.

PAGE 48

48 CHAPTER 4 COMPUTED TOMOGRAPHY DOSE PROFILES Introduction Thus far, fiber-optic coupled dosimeters have been characterized for their performance across the diagnostic energy range. Their sensitivity and reproducibility suggest they may be useful in measuring fundamental parameters of radiation dosimetry, namely the direct recording of the radiation characteristics associated withdiagnostic imaging modalities. The real-time response of these dosimeters furthersuggests their value in characterizing the dose delivery methods of multi-detector computed tomography, where ion chamber collected charge measurements fail and the use of thermo-luminescent dosimeters can be extremely time consuming. Therefore, comprehensive analyses and measurements of computed tomography dose profiles have been performed using a fiber-optic coupled dosimetry systemto characterize the various operating conditionsof a computed tomography scanner. Materials and Methods Helical CT scans were performed usinga Siemens Somatom Sensation 16-slice scanner. Dose measurements were recorded using FOC dosimeters and polymethylmethacrylate(PMMA) CT dosimetry phantoms. Table 4-1 lists the combination ofcomputed tomography operating conditions utilized, along with the types of CTDI phantomschosenand the positions used within each phantom. For each measured dose profile, the active area of the dosimeterwasplaced within the center of either the head or body phantoms CTDI positionand the counts were measured versustime. The helicalscan dose profiles were then plotted as counts versus position within the phantom(i.e. distance along the z -axis).

PAGE 49

49 Table 4-1.Helical scan dose profiles Operating conditions. Scans were performed using the five positionsfor two CTDI phantoms, at three pitches and two slice thicknesses. phantom kVp effective mAs position pitch slice thickness a Head 120 82 Top 0.5 12 mm b Body 120 133 Bottom 1 24 mm c Center 1.5 Left Right a Slice thickness, T = n T where n T = beam width b Acquired using T = 16 x 0.75 mmc Acquired using T = 16 x 1.5 mm Figure 4-1. Photograph of the standard PMMA CTDI phantoms. A) Body. B) Head. A B

PAGE 50

50 Head & Body Phantoms Measurements were performedusing twostandard PMMA CTDI phantoms: the head and the body phantoms. Bothphantomsare 15 cm in length. The head phantom is 16 cm in diameterand the larger body phantomhas a diameter of 32 cm. In each of the CTDI phantoms there are five drilled holes (1 cm in diameter) parallel to the z -axis (couch direction) to allow the placement of dosimeters for the measurement of the central and peripheral CTDI. Holes that are not being used are filled with corresponding PMMA plugs. Figure 4-1 shows the two phantoms and the five drilled holes/positions (center, top, left, right, and bottom). Tube Potential (kVp) & Tube Current-Time Product (mAs) Dose profiles were measured using five different positions within both the head and body phantomsfor a permutation of three pitches and two slice thicknesses (i.e. a total of 30 head phantoms scans and 30 body scans). All scans were performed at a tubepotential of 120 kVp. The tube current-time product(mAs) was 82 and 133 effective mAs for the series of head and body phantom scans, respectively(Table 4-1). Tube current and exposure time per rotation govern the number of x-ray photons utilized, givenby the tube current-time product (mA s), or simply mAs (milliamperes-second). Tube current-time product is indicative of the relative output (radiation exposure) of thex-ray tube of a given CT scanner, at a given kVp. It does not indicate the absolute output (dose), as the exposure per mAs varies significantly between CT scanner manufacturers, models, and tube potentialsettings.1 Siemens reports the average mAs along the z-axis, called the effective mAs, defined as the true mAs divided by the pitch. In MDCT, noise is dependent on pitch, thus as pitch is increased, scanner software may automatically increase the mA such that image noise (and possibly the

PAGE 51

51 patient dose)remains relatively constant with changing pitch values.5,41 Thus, the user may be unaware that the actual mA was increased. Pitch For the Siemens Somatom Sensation 16-slice scanner, pitch is defined as the table increment per rotation of the gantry divided by the detector acquisition width. 42 It is an important aspect of the scan protocol, influencing the dose to the patient,scan time, and image quality. Measurements in this study were performed using three different selections of pitch: 0.5, 1.0, and 1.5. A pitch of 1.0 is physically similar to performing a conventional axial scan with contiguous slices. 12 Scanning with a pitch of less than 1.0 results in overscanning, or an overlapping of single scan dose profiles. Its advantage is an improvement in image quality, but it comes at the cost of higher dose to the patient. On the other hand, increasing the pitch beyond 1.0, leads to a partial scanning of the patient and a potential reduction in image quality. The advantage of increasing pitch for a given CT protocol is both a reduction in scan time as well as a reduction of dose to the patient if all otherparameters are held constant. Slice Thickness The slice thickness T for multi-detector CT scanners is determined by the number of channels N acquired along the z T T = n T (4-1) For example, the 12 mm slice thickness (nominal beam width) used in measuring the helical dose profiles below is acquired using T = 16 0.75 mm. The detector configuration not only determines the desired slice thickness, but also affects the retrospective reconstruction options (thinner or thicker images) and the radiation efficiency of the system (i.e. patient dose). 1

PAGE 52

52 Measurements in this study were performed at two different nominal slice thicknesses: (1) 12 mm (acquired at T = 16 0.75 mm) and (2) 24 mm (acquired at T = 16 1.5 mm). The wide beam collimation allows for much faster z -coverage (i.e. faster scan times). The narrower collimation allows for retrospective reconstruction of narrower slices. Thinner slices have the benefit of improved spatial resolution with a trade-off of increased dose to the patient. Helical scan dose profiles were performed using various slice thicknesses and pitches for both the head and body phantoms at the centers and peripheries of each phantom. The active area (phosphor) of the fiber-optic coupled dosimeter was placed in the center of the selected phantom position and scans were performed for the various clinical protocols listed inTable 4-1. The fiber response (counts) was measured in real time and plotted as counts versus phantom position. The characteristics of the dose profiles were then analyzed in terms of changes in CTDI phantom position, pitch, and slice thickness. Resultsand Discussion Head Phantom Figure 4-2shows a comparison of measured helical scan dose profilesperformed in the five positionsof the CTDI head phantom. Each profileshown was performed at 120 kVp and 82 effective mAs for a pitch of 0.5 and aslice thicknessof 12 mm. The accuracy of the dosimeters real-time acquisition response was verified by comparingthe photon counting heads time readoutversus the total scan time displayed on the Siemens CT console. Figure 4-2 displays the intensity response ofthe fiber plotted versus position in the PMMA phantom. Fiber response (counts) is normalized to the maximum dose D (0) which occurs at the center of the scan length.

PAGE 53

53 Figure4-2. Comparison of measured datain the five positionsof the CTDI head phantom. Center and four peripheral(left and right responses are identical)measurement locationsindicated. Profiles were measured from CT scans performed at 120kVp and 82 effective mAs for selected pitch of 0.5 and aslice thicknessof 12 mm. The first characteristic one notices is the oscillating nature of the helical dose profiles. Figure 4-3 provides an expanded view of the dose profile seen in the top left of Figure 4-2(note the abscissa change to units of CT scan time). The tube rotation time of 0.5 seconds can be seen from the measured distance between adjacent peaks in the dose profile. Also shown are the attenuation effects of the CTDI phantom material as the tube rotates around the phantomdosimeter setup. In Figure 4-3, the FOC dosimeter is placed in the center of the top position of the CTDI phantom. Peaks in the dose profile correspond to an x-ray tube position of 12 oclock Top Left/Right Bottom Center

PAGE 54

54 (i.e. shortest possible source-to-detector distance, minimal x-ray attenuation). Minima in the profile are due to the location of the tube at 6 oclock(the longest source-to-detector distance). The combined effects of the CT table and 16 cm diameter head phantom result in an attenuation of 99.3% of the dose when the x-ray tube is at a maximum distance from thedosimeter. Figure 4-3. Expanded view of thehelical dose profile measured in the 12 oclock(top) position of theCTDI head phantom. The CT scan was performed at 120 kVp and 82 effective mAs for selected pitch of 0.5 and a slice thickness of 12 mm. Theattenuation effects of both the PMMA phantom material and the CT table are also visible in the other dose profiles of Figure 4-2. The head phantoms bottomposition dose profile has a lower intensity (85%) at the center of the scan length when normalizedtothe other peripheral dose profiles (top, left, and right) due to theincreasedx-ray attenuation from the CT table. The peaks in the bottom positionprofile occurwhen the x-ray tube is at 6 oclock. Here the source-to-detector distance is minimized,but reduced intensity in the center dose is caused by the attenuation of the table. Correspondingly, the attenuation effects at the maximum source-

PAGE 55

55 to-detector distance for the bottom phantom position are not as significant. When the tube is located at 12 oclock, the phantom materialalone contributes to the attenuation of the dose measured by the dosimeter. Without the addition of the CT table, 86.3% of the central dose is attenuated when the x-ray tube is at a maximum distance from the dosimeter. Whenthe dosimeter is placed in the center position of the head phantom for the measuring of helical dose profiles, the attenuation effects of the PMMA phantom are constantduring the rotation of the x-ray tube. As a result, there is alowering of the dose distribution at the central slice of the centerposition profile (bottom rightof Figure 4-2). Figure 4-4is an expanded view ofthis profile with the abscissa changed to reflect the helical scan time. The tube rotation time of 0.5 seconds can be seen in the measured distance between consecutive minima in the dose profile, which correspond to an x-ray tube position of 6 oclock. Figure 4-4. Expanded view of thehelical dose profile measured in the center position of the CTDI head phantom. The CT scan was performed at 120 kVp and 82 effective mAs for selected pitch of 0.5 and a slice thickness of 12 mm.

PAGE 56

56 A close look at the peak dose at the center of the scan length shows a flattening of the peak dose when the tube is above the table. This flattening of the peak is due to the symmetry of x-ray attenuation from the cylindrical phantom. Also seen in the zoomed view of Figure 4-4 is a small dip in the central dose peak measured to be 0.25 seconds from adjacent minima, thus corresponding to a tube position of 12 oclock. The difference in intensity between peak and valley, therefore, is due solely to the CT table which attenuates 33% of the dose at isocenter. Figure 4-4 illustrates the significance of table attenuation in the dose profile not seen in other models that take into account only phantom attenuation in the axial plane. Body Phantom Measured helical scan dose profiles were also performed for the five positions of the CTDI body phantom for three pitches and two slice thicknesses. Figure 4-5displays the normalized intensity response of the fiber plotted versus position in the 32 cm diameter PMMA body phantom. The profiles were measured from CT scans performed at 120 kVp and 133 effective mAs for a 12 mm slice thickness and pitch of 0.5. The most obvious characteristic of the body phantom dose profiles is the attenuation effects of the larger diameter phantom (compared to the smaller 16 cm diameter head phantom). Note the intensity difference between the peaks and valleys for each peripheral dose profile. The 32 cm diameter body phantom attenuates nearly 100% of incident x-rays when the tube is directed from the opposite side of the dosimeter (maximum source-to-detector distance). Thus, x-ray attenuation in the body phantom is dominated by the large diameter phantom and not the CT table. The head phantoms bottom position dose profile has a lower intensity (86%) at the center of the scan length when normalized to the other peripheral dose profiles (top, left, and right) due to the increased x-ray attenuation from the CT table. The effect of table attenuation on the central peak of the body phantoms bottom position profile is not as significant

PAGE 57

57 as in the head phantom, though the intensity of the dose at the center of the scan length is still lower when compared to the other peripheries (91%). Figure 4-5. Comparison of measured datain the five modules of the CTDI bodyphantom: center and four peripheries. Profiles were measured from CT scans performed at 120 kVp and 133 effective mAs for a pitch of 0.5 and 12 mm slice thickness. The lowering of the central peak dose in the center position of the body phantom dose profile is also enhanced due to the increased x-ray attenuation of the larger diameter body phantom. The broadening of the scatter tails in the center position dose profile can be seen in the lower right graph ofFigure 4-5, and is also due to the larger diameter body phantom, as the scatter-to-primary ratio increases with increasing phantom diameter thickness. The broad scatter tails for the body phantoms center position fall off sharply to zero at the end of the phantom Top Left/Right Bottom Center

PAGE 58

58 length, indicating that standard PPMA phantom lengths (150 mm) are too short to sufficiently measure the scatter radiation associated with helical computed tomography. Pitch Head phantom Helical CT scans were also analyzed to illustratethe effect of pitch selection on the measured dose profile. Again, the active area of the dosimeter was placed in the center of the selected CTDI phantom positionand helical scans were produced. Doseprofiles in Figure 4-6 were measured at 120 kVp, 82 effective mAs, anda slice thickness of 24 mmin the top and center positions of a CTDI head phantom. The pitch associated with each scan is indicated within the figure along with the phantom positionand scan time. Figure 4-6 shows the change in the dose profile as pitch is increased and slice thickness is held the same. The reduction of scan time was indicated on the CT console and confirmed by the dosimetry system. The integrated counts (dose) were also reduced as pitch increases for the five positionsof the head phantom(Table 4-2). Compared to the measured data for a pitch of 1.0, the total counts (integrated across the entire scan length) for all head phantom position dose profiles show a 15 and 20% increase in dose for the 12 and 24 mm slice acquisition, respectively, when the pitch was reduced to 0.5. Likewise, an increase in pitch from 1.0 to 1.5, leads to an average dose reduction of 9 and 13% for the two slice thicknesses in the five CTDI position dose profiles. The weighted average given in Table 4-2takes into account the fact that dosedistribution varies in the axial (xy) plane given by CTDIW (Eq. 1-2). CTDI (W) = 3 1 CTDI (C) + 3 2 CTDI (P) (4-1) where CTDI (P) is the average of the four peripheries.

PAGE 59

59 Figure 4-6. Effect of pitch on measured helical dose profiles in the CTDI head phantom. Profiles were measured from CT scans performed at 120 kVp, 82 effective mAs, and 12 mm slice thickness. Phantom position, pitch, and scan time are indicated. The effective mAs in Siemens MDCT, defined as the true mAs divided by the pitch, is used so that scanner software will automatically increase the mA such that image noise (and possibly the patient dose) remains relatively constant with changing pitch values. However, dose is still seen to increasewith decreases in pitch, and vice versa (Tables 4-2 and 4-3). An evaluation of the efficiency of the CTDI metric (Chapter 5) shows that the primary (central peak dose) dose delivered to a cylindrical PMMA phantom (regardless of position) does not change significantly when pitch is changed. Therefore, the difference in dose ratiosis due primarily to Top 0.5 pitch Scan time 14.94s Top 1.0pitch Scan time 7.86s Center 0.5 pitch Scan time 14.94s Center 1.5 pitch Scan time 5.51s

PAGE 60

60 scatter radiation outside the center slice. When pitch is reduced,overscanningof the phantom allows for the measurement of the scatter tails at the edges of the phantom length. Table 4-2.Head phantom -Ratios of integrated counts (dose) show the effect of pitch on the helical scan dose profile. FOC dosimeters measured the integrated counts from helical dose profiles acquired at 120 kVp and 82 effective mAs in five positions of a CTDI head phantom for two slice thicknesses and three pitches (0.5, 1.0, 1.5). slice thickness 12 mm 24 mm Pitch Ratio module 0.5 / 1.0 1.5 / 1.0 0.5 / 1.0 1.5 / 1.0 top 1.17 NA 1.22 0.87 left 1.16 0.89 1.23 0.87 right 1.15 0.91 1.20 0.86 bottom 1.13 0.92 1.17 0.87 center 1.13 0.91 1.18 0.89 average ( W ) 1.14 0.91 1.20 0.88 Body phantom Figure 4-7 shows the change in the body phantom helical scan dose profiles as pitch is increased. The scan time is indicated next to each dose profile and shows a reduction in total scan time as pitch is increased. The selected pitch, phantom position, and slice thickness arealso shown within the figure. Table 4-3 lists the ratio of integrated counts (dose) as pitch is changed and slice thickness is held the same for all five positions of the CTDI body phantom. The periphery ratios are similar to calculated head phantom data, with the exception of the body phantoms center position. Nevertheless, the weighted average remains close to that calculated in the head phantom due to the heavy weighting given to the peripheral points in the axial slice plane, where primary-to-scatter ratios are higher than in the center position.

PAGE 61

61 Figure 4-7. Effect of pitch on measured helical dose profiles in the CTDI body phantom. Profiles were measured from CT scans performed at 120 kVp and 133 effective mAs. Phantom position, pitch, slice thickness, and scan time are indicated. The total counts integrated across the entire scan length (i.e. dose) are reduced as pitch is increased for the four peripheries, yet the dose in the center position of the body phantom shows little change as pitch is increased or decreased (Table 4-3). This is due to the increased scatterto-primary ratio of incident radiation at the center position of the larger diameter phantom and the finite size of the phantom length. The lowering of the central peak dose in the center position of the body phantom profile is due to the increased x-ray attenuation of the larger diameter body phantom. The broadening of the scatter tails in the center position dose (bottom graphs of Figure4-7) is also due to the larger Bottom 24 mm slice 0.5 pitch Scan time 8.81s Bottom 24 mm slice 1.0pitch Scan time 4.65 Center 12mm slice 0.5 pitch Scan time 14.94s Center 12mm slice 1.5 pitch Scan time 5.51s

PAGE 62

62 diameter body phantom, as the scatter-to-primary ratio increases with increasing phantom diameter thickness. As pitch is changed, the primary (central peak dose) dose delivered to the body phantom does not change significantly. Therefore, the difference in dose with change in pitch is due primarily to scatter radiation outside the center slice. The broad scatter tails for the body phantoms center position fall off sharply to zero (regardless of pitch selection, Figure 4-7) at the end of the phantom length, indicating that standard PPMA phantom lengths (150 mm) are too short to sufficiently measure the scatter radiation associated with helical CT. Table 4-3.Body phantom -Ratios of integrated counts (dose) show the effect of pitch on the helical scan dose profile. FOC dosimeters measured the integrated counts from helical dose profiles acquired at 120 kVp and 133 effective mAs in five positions of a CTDI body phantom for two slice thicknesses and three pitches. slice thickness 12 mm 24 mm Pitch Ratio module 0.5 / 1.0 1.5 / 1.0 0.5 / 1.0 1.5 / 1.0 top 1.18 0.88 1.28 0.86 left 1.17 0.90 1.22 0.88 right 1.16 0.92 1.26 0.87 bottom 1.15 0.91 1.17 0.87 center 1.03 0.97 1.10 0.94 average (W) 1.12 0.92 1.19 0.89 Slice Thickness Head phantom Figure 4-8 shows the change in helical dose profiles as the detector configuration (slice thickness) is changed and pitch is kept the same. As expected, the scan time is greatly decreased when a wider detector configuration is chosen and all otherparameters remain unchanged. The dose integrated across the scan length is reduced as well. Table 4-4 displays the ratio of

PAGE 63

63 integrated counts measured with fiber-optic coupled dosimeters as the slice thickness is reduced from 24 to 12 mm. Figure 4-8. Effect of slice thickness on measured helical dose profiles in the CTDI head phantom. Scans wereperformed at 120 kVp and 82 effective mAs. Phantom position, pitch, slice thickness, and scan time are indicated Integrated counts were measured from helical dose profiles acquired at 120 kVp and 82 effective mAs in a CTDI head phantom for two slice thicknesses. The integrated count ratio of narrow versus wide slice thickness was then calculated for the five CTDI positions and three pitches. As stated earlier, narrower collimation allows for retrospective reconstruction of narrower slices, which have the benefit of improved spatial resolution with a trade-off of increased dose to Right 1.0 pitch 12 mm slice Scan time 7.86s Right 1.0 pitch 24mm slice Scan time 4.65s Center 0.5 pitch 12 mm slice Scan time 14.94s Center 0.5 pitch 24 mm slice Scan time 8.81s

PAGE 64

64 the patient. The dose profiles for each phantom position show thisincrease in dose as the slice thickness is narrowed. At a pitch of 0.5, the measured counts increase by an average of 20% as slice thickness is reduced from 24 to 12mm. The increase in counts average 25% for a pitch of 1.0 and was 30% for a pitch of 1.5. Table 4-4.Head Phantom -Ratios of integrated counts (dose) show the effectof slice thickness on the helical scan dose profile. FOC dosimeters measured the integrated counts from helical dose profiles acquired at 120 kVp and 82 effective mAs inthe five positions of a CTDI head phantom for two slice thicknesses and three pitches. 12 mm slice / 24 mm slice pitch module 0.5 1 1.5 top 1.22 1.27 NA left 1.18 1.25 1.27 right 1.20 1.25 1.32 bottom 1.19 1.24 1.31 center 1.19 1.25 1.28 average (W) 1.20 1.25 1.30 Body phantom Figure 4-9shows the change in helical dose profiles as the slice thickness is changed. Again, the scan time and integratedcounts are decreased when a wider detector configuration is chosen and all other parameters remain unchanged. Integratedcounts were measured from helical dose profiles acquired at 120 kVp and 133effective mAs in theCTDI bodyphantom. The ratio of integrated counts (dose) of narrow versus wide detector configurations was then calculated for the five CTDI modules atthree pitches. Table 4-5 shows the ratio of integrated dose as the nominal slice thickness is changed from 12 to 24 mm. As expected, the top, left, and right phantom position dose ratiosfor the body phantomshow a steady increase in dose of27, 36, and 40%,for pitches of 0.5, 1.0, and 1.5,

PAGE 65

65 respectively, when slice thickness is reduced from 24 to 12 mm. A closer look at Tables 4-4 and 4-5 shows that body phantom ratios (at thetop, left, and right) are slightly higher than corresponding dose ratios in the head phantom. Figure 4-9. Effect of slice thickness on measured helical dose profiles in the CTDI body phantom. Scans wereperformed at 120 kVp and 133 effective mAs. Phantom position, pitch, slice thickness, and scan time are indicated. This is due to primarily to the detectors proximity to the radiation source and the added attenuation of the larger diameter phantom. For peripheral phantom positions (excluding the bottom), the larger diameter body phantom places the dosimeter 8 cm closer to the x-ray tube when the source-to-detector distance is minimized. As seen in Figure 4-5, the 32 cm diameter body phantom attenuates nearly 100% of incident x-rays when the tube is directed from the Top 12mm slice 0.5 pitch Scan time 14.94s Top 24mm slice 0.5 pitch Scan time 8.81s Left 12mm slice 1.0pitch Scan time 7.86s Left 24mm slice 1.0pitch Scan time 4.65s

PAGE 66

66 opposite side of the dosimeter (maximum source-to-detector distance). Therefore, dose ratios in the body phantom with regard to change in slice thickness are predominately dependent on the primary radiation when the source-to-detector distance is minimized. Table 4-5.Body Phantom -Ratios of integrated counts (dose) show the effect of slice thickness on the helical scan dose profile. FOC dosimeters measured the integrated counts from helical dose profiles acquired at 120 kVp and 133 effective mAs in the five positions of a CTDI body phantom for two slice thicknesses and three pitches. 12 mm slice / 24 mm slice pitch module 0.5 1 1.5 top 1.26 1.38 1.40 left 1.29 1.34 1.37 right 1.25 1.35 1.42 bottom 1.23 1.26 1.31 center 1.18 1.27 1.31 average (W) 1.23 1.31 1.35 The body phantoms center and bottom ratios show an increase in dose (21, 27, and 31% for pitches of 0.5, 1.0, and 1.5, respectively) with decreasing slice thickness, but at a lesser rate than the other peripheries. This is due to the increased scatter-to-primary ratio (SPR) at these two phantom positions. For the center phantom position, the SPR is high relative to the peripheries due to theincrease in phantom attenuation. At the bottom position, scatter contributions are increased and the primary beam is attenuated by the CT table. In comparing head and body phantom data, the ratios are similar since the minimal source-to-detector distance is the same for both bottom positions and the only difference is the limited dose attributed to scatter when the source-to-detector distance is at a maximum (nearly negligible for both).

PAGE 67

67 Conclusions Fiber-optic coupled dosimeters,based on the sensitive elements of coupled scintillation phosphors have been implemented in measuring the dose delivery methods of multi-detector computed tomography. Comprehensive analyses of computed tomography dose profiles have been performed using a fiber-optic coupled dosimetry system and standard PMMA dosimetry phantoms forvarious operating conditions of a Siemens Somatom Sensation 16-slice scanner. FOC dosimeters sensitivity and reproducibility in the diagnostic energy range, along with their ability to directly measure the primary and scatter radiation associated with helicalMDCT scanning, provide alternative methods for determining CT dose profilesand allow for the evaluation of CT dose characteristics not commonly observed by standard dosimeters.CTDI Phantom Position The effect of phantom position choice on the intensity of the measured dose profile fits well with measured single scan dose profile results performed by Tsai et al using LiF TLDs and a GE ProSpeed CT scanner. 22 However, helical scan dose profiles measured with fiber-optic coupled dosimeters provide information of CT delivery characteristics that more conventional computed tomography dosimetry metrics lack. The small size of the fibers active area allows for accurate point dose measurements. The fiber-optic dosimetry system allows for remote, real-time dose measurements. The direct recording of the helical scan profiles that characterizemulti-detector CT can tell us more about the dose delivery of a given CT system than could a collected charge measurement from a standard pencil ion chamber. The fibers real-time response also saves valuable time during measurements without the annealing of thermoluminescent dosimeters. Helical dose profiles measured with FOC dosimeters and plotted versus timeand/or phantom position portray the periodic motion of helical scanning from above to below the table

PAGE 68

68 top, clearly seenin the peaks and valleys of the measured dose profilesas the scanner cycles around the dosimeter. Peaks in the dose profile correspond to an x-ray tube position closest to the detector and corresponding minima occur when the source-to-detector distance (and associated phantom and/or table attenuation) is at a maximum. Dose profiles acquired in the center position of CTDI phantoms show both a lowering and flattening out of the central dose peak due to the symmetry of x-ray attenuationfrom the cylindrical phantom and the constant source-to-detector distance. The scatter tails not measured by conventional CTDI 100 are also seen in helicaldose profilesthat scanthe entire length of the phantom (Chapter 5).Head vs. Body Phantom Helicalscan dose profiles displaythe attenuation effects of PMMA material and the CT tableitselfwhen fiber-optic dosimeters are placed in the various positionsof the CTDI phantom. The top, left, and right periphery fiber responses are similar for profiles measured in the head phantom, while reduced intensities of the measured dose profiles central doseare apparent for the bottom peripheries due to x-ray attenuation from the CT table. The center positiondose profiles show a further increase inattenuation of x-rays due to the phantom material of the 16 cm diameter head phantom. Body phantom datashowsan even more significant reduction of dose at the center of the scan lengthin the measured profile as a result of the increasedamount of phantom material when a larger diameter phantom is chosen. The lowering and flatteningof thecentral dose and the broadening of the scatter tails inthe central position dose profile is also enhanced due to the increased x-ray attenuation and the higher scatter-to-primary ratio of the larger diameter body phantom. The top, left, and right periphery profiles show similar responseswith respect to the peak dose at the center of the scan length and the intensity differencesbetween the peaks and valleys.

PAGE 69

69 The 32 cm diameter body phantom attenuates approximately 100% of incident x-rays when the tube is at a maximum distance from the dosimeter. The effect of table attenuationon the dose profile of the bottom phantom position is not as significant as in the head phantom(86%), though the intensity of the dose at the center of the scan length is still lower when compared to the other peripheries (91%). Pitch The measurement of single scan dose profiles only provides information regarding a single axial slice. Thus, the effect on changes in dose delivered due to a change in pitch in contiguous scanning cannot be determined fromconventionalSSDPs. However, helical scan dose profiles measured with fiber-optic coupled dosimeters can provide information regarding the characteristic effect of pitch on a CT systems dose delivery. Measured data shows the increased dose typical of overscanning whenscanner pitch is less than 1.0, despite the scanners use of effective mAs to maintain dose with changes in pitch. Scan time of measured dose profiles increased at lower pitches and integrated dose increased accordingly with an average of 15-20% and 17-25% for the head and body phantoms, respectively. As pitch was increased from 1.0 to 1.5, measured dose profiles for the five CTDI positions showed an average decrease in dose of 9-13% for both the head and body phantoms. The following chapter willshow that the primary (central peak dose) dose delivered to a cylindrical PMMA phantomdoes not change significantly when pitch is changed. Therefore, the difference in dose ratios is due primarily to scatter radiation outside the center slice. When pitch is reduced, overscanningof the phantom allows for the measurement of the scatter tails at the edges of the phantom length and thus leads to an increase in dose.

PAGE 70

70 Slice Thickness As expected, the top, left, and right position dose ratios for the body phantom show a steady increase in dose of 27, 36, and 40%, for pitches of 0.5, 1.0, and 1.5, respectively, when slice thickness is reduced from 24 to 12 mm. These values are slightly higher than corresponding dose ratios in the head phantom due primarily to the detectors proximity to the radiation source and the added attenuation of the larger diameter phantom. The body phantoms center and bottom ratios show an increase in dose with decreasing slice thickness, but at a lesser rate than the other peripheries. For the center phantom position, the scatter-to-primaryratio (SPR) is high relative to the peripheries due to the increase in phantom attenuation of the large diameter body phantom. At the bottom position, scatter contributions are increased and primary radiation is attenuated by the CT table. In comparing head and body phantom data, the ratios are similar since the minimal source-to-detector distance is the same for both bottom positions and the only difference isthe limited dose attributed to scatter when the source-to-detector distance is at a maximum (nearly negligible for both).

PAGE 71

71 CHAPTER 5 COMPUTED TOMOGRAPHY DOSE INDEX Introduction Analyses of single scan dose profiles measured with standard pencil ion chambers illustrate the shortfallof traditional CTDI toaccurately predict the dose delivered from helical image acquisitions, by failing to sufficiently measure the scatter tail radiation characteristic of multi-detector computed tomography. The use of polymethylmethacrylate CTDI phantoms along with fiber-optic coupled dosimeters have shown potential inthe direct measurementofthe primary and scatter radiation associated with multi-detector, helical CTscanning. Fiber-optic dosimeters ability to both discriminatethe scatter tails associated with helical CT and measure the dose associated with scan lengths longer than the 100 mm of pencil ion chambers provide alternative methods for determining CT dose profiles. CTDI 100 The computed tomography doseindex (CTDI) was originally defined in order to characterize the radiation dose properties of single slice CT scanners. It is traditionally measured usinga 100 mm long pencil ion chamber with 150 mm long, cylindrical polymethylmethacrylate (PMMA) dosimetry phantoms. Despite advancements in CT technology, CTDI remains a widely used metric for the dose performance of multi-detector CT(MDCT) and cone-beam CT (CBCT) scanners. 43-44 Recently, however, the accuracy of the CTDI metric for patient dosimetry with modern multi-detector CT scanners has come into question. 25-26,45-47 Pencil Ion Chamber The limitations of the computed tomography dose index-100 can be seen in its definition CTDI 100 = T 1 mm mm 50 50 f( z) dz (5-1) where f(z) is the single, axial scan dose profile along a line parallel to the z -axis of the scanner

PAGE 72

72 for nominal slick thickness T If the scan length L = 100 mm, the10 cm long pencil ion chamber will predict the maximum dose D (0) at the center ( z =0) of the single scan profileas defined in Chapter 1. For shorter or longer scan lengths this is not necessarily true. 26 As the cumulative dose profile D(z) (Eq. 1-6) builds up from the integration of single scan profiles f(z) the dose in the central region flattens out and reaches an equilibrium value D eq (0)(Eq. 1-7) when L is large enough to encompass the scatter tails of f(z) 26-27 If the center dose reaches equilibrium for L 100 mm, then the 10 cm long chamber will predict both D eq (0) and DLI(Eq. 1-8). If equilibrium is not reached, however, these quantitieswill be underestimated. 26 Small Volume Ion Chamber Scan lengths utilized by most clinical exams are long enough such that the dose equilibrium should be reached in the center of the scan. Thus, the equilibrium dose can be measured if the chamber is longenough to encompass all the scatter tail radiation resulting from the single scandose profile. Instead of making the ion chamber longer, Dixonet al.proposed an alternative to the CTDI method,suggestingthe use of a small volume ion chamber to scan a lengthof phantomlong enough to establish dose equilibrium at the location of the chamber.26,48 The use of a small volume ion chamber to directly measure the cumulative dose D(x) at any point by scanning a length of phantom long enough to produce dose equilibrium atthe center of the scan lengthhas the same result as making the chamber longer and is shown to bemore accurate for wide beam profilesand scan lengths greater than 100 mm. Even if the scan length L is not long enough to produce equilibrium at the center, such a small chamber will give a good measurement of the maximum doseat the center, whilea 10 cm chamber will merely give the average dose over the central 10 cm of the scan length.

PAGE 73

73 Dixon and Ballard experimentally implementedthe small ion chamber method by measuring the accumulated dose in a 400 mm long, 32 cm diameter PMMA body phantom for variousscan lengthsL includingtheequilibrium dose D eq (or CTDI ). 49 Their results showed good agreement in accumulated dose values (%)between small ion chamber and pencil ion chamber measurements when the scan length equaled the length of the pencil ion chamber, but failedto show similar correlation at more clinically relevant scan lengths. The measured equilibrium doses obtained by Dixon and Ballard at a scan length of 400 mm for GE MDCT scanners at 120 kVp were CTDI = 1.75 CTDI 100 on the central axis and 1.22 CTDI 100 at the peripheries. 49 Nakonechny et al showed that for nominal beam widths ranging from 3-20 mm and for scan lengths of 250 mmaccumulated dose values at the center phantom position were approximately 25-30% higher than the measured CTDI 100 Peripheral point measurements were less severe, but differences were as much as 22%. 47 Theauthorsmeasured SSDPs with PTW diamond detectors, lithium fluoride TLDs, and asmall volume ion chamber. The profiles were acquired in elliptical water-equivalent phantoms (major andminor axes of 30 and 20 cm, respectively, and 30 cm in length) and the relative accumulated dose was measured at the center for various scan lengths L The accumulated dose reached equilibrium for L > 300 mm(in agreement with Dixonet al.), suggesting the need for phantoms longer than standard CTDI phantoms. 47,49 CTDI 100 measurements were also made using the small ion chamber and were within 4% of a 102 mm length pencil ion chamber for a scan length of L = 100 mm, further suggesting that dosimetersother than the standard 10 cm pencilion chamber can besuccessful in clinicalCT dosimetry.47

PAGE 74

74 Material and Methods Fiber-Optic Coupled Dosimeters Helical scan dose profiles measured with fiber-optic coupled dosimeters have shown potential in the characterization of multi-detector computer tomography dose delivery. Their ability tocollect thescatter tails associated with helical CT suggests promise in measuringthe accumulated doseof clinically relevant CT scan lengthswhere the use of standard pencil ion chambers fail. Chapter 4 described the use of PMMAphantoms andFOCdosimeters to measure helical scan dose profiles. The following sections analyze these dose profiles and illustrate the shortfall of CTDI 100 in accurately predicting the dose delivered from helical image acquisitions. CTDI Efficiency As previously discussed, recent studies have called into question the accuracy of the computed tomography dose index for patient dosimetry. John Boone 25 proposedan evaluation of the efficiency of the CTDI 100 metric using Monte Carlo simulation techniques. The CTDI efficiency was definedas the ratio of the average dose (accrued at z =0) for multiple contiguous slices over a scan length of 100mm,tothe equilibrium dose accrued at z =0, which is approached asymptotically as the scan length L becomes wider than the single scan dose profile, D(z) to include the scatter tails (as L 25 z mm z mm dz z D nT dz z D nT ) ( 1 ) ( 1 50 50 = CTDI CTDI 100 (5-1) where D(z) is the dose deposited for the single scan profile along the z -axis. The Monte Carlo simulationsutilized the geometry of a commercially available CT scanner with a modeled polyenergetic x-ray spectra. Infinitely long head (16 cm diameter) and body (32 cm) PMMA phantoms were modeled and dose spread functions (DSFs) were computed

PAGE 75

75 along the length of 12.4 mm diameter PMMA rods placed at radii reflecting the five positions of standard CTDI phantoms. The CTDI 100 efficiency was then calculated as the fraction of the dose along a PMMA rod collected in a 100 mm length scan centered on the CT slice position, divided by the total dose deposited along an infinitely long PMMA rod. 25 Thismethod was adopted tocalculate the efficiency of CTDI100 with respect to the helical scan dose profiles (Chapter 4) measured with fiber-optic coupled dosimeters at the five positions ofstandard PMMA head and body phantoms undervarious operating conditions. The CTDI100 efficiency was determined as the ratio of the accumulated dose integral measured in the center 100mm of the phantom divided by the total 150mm scan length of the PMMA phantom that includes the scatter edges. 150 mm mm CTDI CTDI 150 100 (5-2) Need for Longer Phantoms Helical dose profiles have been measured at the center and peripheral locations of CTDI phantomssincecomputed tomography dose varies across the field of view. The weighted CTDI (discussed in Chapter 1) takes into account the fact that dose is distributed unevenly in the axial ( xy) plane. This is especially true for the center position of the larger diameter body phantom, as the scatter-to-primary ratioincreases with increasing phantom diameter thickness50 (see Figure 4-5 and Tables 4-3,4-5). WhereCTDIW representsthe average absorbed dose in the x-y plane, the volume computed tomography dose index (CTDI vol )estimates the average radiation dose within the irradiated volumeof the CTDI phantom. It does not, however, represent the average dose for objects of substantially different size, shape, attenuation or when the 100 mm integration limits omit a considerable fraction of the scatter tails. 25

PAGE 76

76 Recent studies implementing the use of a small ion chamber to measure the accumulated dose, including the equilibrium dose D eq (0), at the center of various scan lengths have suggested the need for longer phantoms, stating that the equilibrium dose (CTDI ) is reached for scan lengths L > 300 mm. 47,49 Dixonand Ballardsuggested that the commonly used phantom length of 150 mm is too short even for the measurement of CTDI 100 producing an underestimate of 7.3%on the central axis and 1.3% on the peripheral when compared to CTDI100 measurements in a 400 mm long PMMA phantom. 49 Novel method for predicting helical scan dose profiles To evaluate the need for longer phantomsto measure the scatter tails associated with clinically relevant scan lengths, helical dose profiles were acquired using fiber-optic coupled dosimeters and extended length (300 mm) PMMA head (16 cm diameter) and body (32 cm) phantoms. Scans were forperformed for thecombination of scan parameterslisted in Table 5-1, using the same tube potential and effective mAs settings as theprofiles measured in Chapter 4, differing only in the scan time needed (and a result, the dose accumulated) due to the extended length of the double phantoms. Table 5-1.Helical scan dose profiles Operating conditions. Scans were performed using double length ( L = 300 mm) head and bodyCTDI phantoms, at three pitches and two slice thicknesses for the five positions of each phantom. Phantom a kVp effective mAs position pitch slice thickness Head 120 82 Top 0.5 12 mm Body 120 133 Bottom 1 24 mm Center 1.5 Left Right a Double phantom length (L = 300 mm)

PAGE 77

77 To measure helical scan dose profiles in extended length phantoms, two standard CTDI head phantoms (15 cm length, 16 cm diameter) were stacked end to end to achieve a 300 mm phantom length. The same method was repeated for the body phantom. The active area of the fiber-optic dosimeter was placed in the center ( z =7.5 cm) of the first (15 cm length)phantom for each position of the head and body phantoms, respectively, and CT acquisitionswere performed for the entire length of the double phantom ( L = 300 mm). When the dose profile is plotted versus phantom position, the dose peak occursat z = 7.5 cm(Figure 5-1). Figure 5-1. Diagram of the 300 mm length phantom setup. Two 150 mm length phantoms were stacked end to end with the active area of the fiber-optic dosimeter positioned in the center ( z = 7.5 cm)of the first phantom. Positioning the FOC dosimeter at z = 7.5 cm in the 300 mm length phantoms provides a method for predicting the helical dose profiles for both a 150 mm and 450 mm length phantom. The methodology that follows explains a setup to predict two separate phantom length dose L = 300 mm z = 7.5 cm fiber active area (phosphor)

PAGE 78

78 profiles in a single measurement. In the case of the 450 mm length scan, only two15 cm length phantoms are needed instead of three, which is beneficial when using cumbersome 32 cm diameter body phantoms. Thefirst 7.5cm of the 300 mm phantomprofile mirrors the first half of the single phantom profiles measured in Chapter 4 (for equal phantom diameter, location, and scan parameters). Figure 5-2illustratesdose profiles measured in the center position of the CTDI head phantom for a 24 mm slice thickness and pitch of 0.5. The first 7.5 cm of the 300mmphantom dose profile (z = 0 to 7.5 cm) is reflected across the center dose peak and superimposed on the singlephantom dose profile. The two dose profiles show good correlation and are within 1.88% in accumulated dose across a 15 cm scan length. Figure 5-2. Helical scan dose profiles in the center position of CTDI head phantomsmeasured with fiber-optic coupled dosimeters. CT scans were performed at 120 kVp and 82 effective mAs for a selected pitch of 0.5 and a slice thickness of 24mm. Plotted are the dose profiles measured in (1) a single (15 cm length) head phantom and (2) the reflected first 7.5 cm of a dose profile measured in a 300 mm length phantomwith FOC dosimeter placed at z = 7.5 cm.

PAGE 79

79 Figure 5-3 gives another example of the similarities between helical scan dose profiles measured across the length of the single phantom (15 cm) and the reflected first 7.5 cm of the dose profile measured in the 300 mm phantomwith FOC dosimeter placed at z = 7.5 cm. Plotted are the dose profiles measured from CT scans performed at 120 kVp and 82 effective mAs for a selected pitch of 1.0 and a slice thickness of 24 mm in the top (12 oclock) position of the CTDI head phantom. Again, the accumulated dose for the two profiles showed good agreement (within 5.62%). Table 5-2 lists the percent difference in accumulated dose for the two single phantom profiles (measured/reflected) acquired at all slice thickness and pitch combinations for the center and peripheries of the CTDI head phantom. Figure 5-3. Helical scan dose profiles in the 12 oclock position of CTDI head phantoms measured with fiber-optic coupled dosimeters. CT scans were performed at 120 kVp and 82 effective mAs for a selected pitch of 1.0 and a slice thickness of 24 mm. Plotted are the dose profiles measured in a single (15 cm length) head phantom and the reflected first 7.5 cm of a dose profile measured in a 300 mm length phantom with FOC dosimeter placed at z = 7.5 cm.

PAGE 80

80 Table 5-2.Calculated percent difference in accumulated dose across a 15 cm scan length forthe measured single head phantom profile and the reflected first 7.5 cm of the 300 mm length phantomprofile slice thickness 12 mm 24 mm pitch 0.5 1 1.5 0.5 1 1.5 module Percent Difference in Accumulated Dose Profiles Single Phantom / Reflected (z = 0 to 7.5 cm) Double Phantom Center 1.13% 2.71% 2.22% 1.88% 2.25% 2.40% Top 4.24% 4.19% NA 3.07% 5.62% 0.68% Bottom 4.43% 3.88% 9.45% 3.16% 2.95% 3.61% Left 5.37% 7.82% 8.24% 3.20% 3.49% 3.93% CTDI L(mm) Reflectingthe first 7.5 cm of a dose profile measured in anL = 300 mm length PMMA phantom with a FOC dosimeter placed at z = 7.5 cm successfully predicts the dose delivered to a full 15 cm length phantom of the same diameter. Using that methodology, the remaining length of thedose profiles (z= 7.5 to 30 cm) measuredin the 300 mm lengthphantom study were reflected about the dose peak ( z = 22.5 cm) to predict the helical scan dose profile in a 450 mm length PMMA phantom. Implementation of such a protocol reducedthe amount of equipment (two phantoms instead of threein this case) needed to measure the dose in longer length phantoms. The accumulated dose measured from 450 mm length helical dose profiles (reflected 22.5 cm of the double phantom study) was then compared to the integrated dose across the dose profiles acquired in a single (150mm length) phantom. Comparisons were made for both the head and body phantom, respectively, for the various scan parameters listed in Table 5-1. Also analyzed was Dixon and Ballards 49 suggestion that the commonly used phantom length of 150 mm is too short even for the measurement of CTDI 100 The CTDI 100 efficiency in the 150 mm

PAGE 81

81 length was thus reevaluated in comparison with similar CTDI 100 efficiency calculations using the 450 mm length helical scan dose profiles. Results and Discussion CTDI 100 Efficiency Table5-3lists the calculated CTDI100 efficiency for helical scan dose profiles measured with FOC dosimeters in the five positions of standard length (15 cm) PMMA head (16 cm diameter) and body (32 cm) phantoms for two slice thicknesses (12, 24 mm) and three pitches (Table 4-1). The 150 was determined as the ratio of the accumulated dose integral measured in the center 100mm of each respective phantom divided by the total 150mm scan length of the PMMA phantom that includes the scatter edges. The 150 efficiency at the four peripheries (top, left, right, bottom) has been averaged and compared with corresponding values at the center position of the two phantoms. Table 5-3also compares the 150 with respectto changes in slice thickness. Profiles measured at varying pitches (0.5, 1.0, 1.5) for a given slice thickness showed little change as seen inthe measuredvariancefor CTDI efficienciesin Table 5-4. As a result, the 150 efficiencies listed were calculated as the average of measured values at three pitches for each given slice thicknessand phantom position permutation. The efficiencies measured show goodagreement with Boones data at all slice thicknesses for peripheries of the headand bodyphantom. Such agreement is not seen, however, at the center of the two phantoms. Neither is there a decrease in efficiency at larger beam widths for the measured CTDI 100 as predicted from the simulations performed by Boone This is due to the finite size of the phantom resulting in the inability to record the scatter of larger beam widths that extend beyond the limits of the phantom.

PAGE 82

82 Table 5-3. CTDI 100 efficiency( 150 )calculated as the ratio of the accumulated dose measured in the center 100mm of a standard PMMAphantom divided by the total 150mm scan length of thephantom. Efficiencies were determined from helical scan dose profiles acquired at the five locations of standard head and body phantoms for slice thicknesses of 12 and 24 mm. Three scans (for three pitches) were acquired at each combination oflocation and slice thicknessand then averaged. Also listed are Monte Carlo simulations performed by Boone 25 for calculating CTDI 100 efficiency( ) as the ratio of the accumulated dose measured in the center 100mm of an infinitely long PMMA phantom divided by the total dose in the infinite phantom. Center Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm 150 0.92 0.95 0.90 0.95 0.82 a 0.81 b 0.63 a 0.62 b Peripheries avg Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm 150 0.92 0.95 0.92 0.95 0.90 a 0.89 b 0.88 a 0.87 b a Calculated with profiles acquired at slice thickness of 10 mm b Slice thickness of 20 mmIt should be noted that these comparisons are made between the physical measurements performed in this study and idealized Monte Carlo simulations. For the Monte Carlo simulations, the x-ray source was simulated as a point source. Therefore, there is no blurring of the dose distributions due to x-ray beam penumbra. X-ray spectra and bow-tie filters were modeled after GE scanner and the effect of table attenuation on the integral dose profile is neglected. CTDI L ( mm)As previously discussed, reflecting the first 7.5 cmofthe dose profiles measured in the 300 mm length phantomsetup with a FOC dosimeter placed at z = 7.5 cm successfully predicts the

PAGE 83

83 dose delivered to the full 150mm length phantom of the same diameter. The remaining 22.5 cm of the 300 mmphantom profiles (z= 7.5 to 30 cm) were reflected about the dose peak (z = 22.5 cm) to predict the helical scan dose profile in a 450 mm (triple length)PMMA phantom. The accumulated dose measured along the triple phantomhelical dose profiles was then compared to the integrated dose across the dose profiles acquired in the single phantom. Table 5-4.Variance of 150 efficiencies calculated from helical scan dose profiles acquired at three pitches for each combination of slice thickness and phantom position utilized. Scans were acquired at the five positions of both the head and body PMMA phantoms for two slice thicknesses at each position. Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm Variance Center 1.07% 0.48% 2.05% 1.23% Peripheries avg 1.03% 0.90% 1.16% 1.09% Table 5-5 lists the accumulated dose ratios of the 450 and 150 mm length dose profiles (CTDI 450 / CTDI 150 ) for the head (16 cm diameter) phantom. As shown in Table 5-2, accumulated dose valuesfor the single head phantom dose profiles were slightly larger than the corresponding reflected ( z = 0 to 7.5 cm) 150 mm profiles. For comparison, Table 5-6 gives the dose ratios (CTDI 450 / CTDI 150 ) for the reflected triple (450mm length) phantom dose profiles versus the reflected single phantom helical scan dose profiles. The accumulated dose ratios given in Tables 5-5 and 5-6 verify the failure of CTDI measurements, performed in a standard 150 mm length phantom, to sufficiently measure the scatter radiation of clinically relevant scan lengths. Accumulated dose values for the measured single head phantom dose profiles (Table 5-5) were slightly larger than the corresponding reflected ( z = 0 to 7.5 cm) 150 mm profiles (Table 5-6). Even taking the conservative approach,

PAGE 84

84 accumulated dose measuredin the 150 mm length phantom underestimates the scatter radiation of the helical dose profile by an average of 24, 19, 14, and 22% for the center, top, left, and bottom phantom positions, respectively, when compared to the accumulated dose in the 450 mm length profile. Table 5-5.Accumulated dose ratios measured from helical dose profiles of 450 mm and 150 length PMMA head phantoms. Calculated ratios reflect the increased dose due to the accumulation of extended scatter tails associated with longer phantoms. Single phantom doses were calculated from helical dose profiles measured in Chapter 4 while z = 7.5 to 30 cmofthe double phantom study were reflected about the peak dose ( z = 7.5 cm) to predict the helical dose profiles for a 45cm lengthphantom. slice thickness 12 mm 24 mm pitch 0.5 1 1.5 0.5 1 1.5 module CTDI 450 / CTDI 150 Center 1.20 1.24 1.25 1.22 1.24 1.27 Top 1.13 1.16 NA 1.18 1.22 1.28 Left 1.10 1.16 1.22 1.04 1.12 1.18 Bottom 1.21 1.21 1.26 1.20 1.23 1.24 *Calculated from single phantom dose profiles measured in Chapter 4. Table 5-6.Accumulated dose ratios measured from reflected helical dose profiles of 450 mm and 150 length PMMA head phantoms. Calculated ratios showthe increased dose due to the accumulation of extended scatter tails associated with longer phantoms. Single phantom (150 mm length) dose profiles were predicted from the reflected first 7.5 cm ofthe double phantom study, while z = 7.5 to 30 cm was reflected about the peak dose ( z =7.5 cm) to predict the dose profiles for a 450 mm length phantom. slice thickness 12 mm 24 mm pitch 0.5 1 1.5 0.5 1 1.5 module CTDI 450 / CTDI 150 ** Center 1.22 1.27 1.28 1.24 1.27 1.30 Top 1.18 1.21 1.23 1.21 1.29 1.29 Left 1.16 1.26 1.33 1.07 1.16 1.23 Bottom 1.26 1.26 1.39 1.24 1.27 1.28 **Calculated from reflected z = 0 to 7.5 cm for the double phantom dose profile

PAGE 85

85 The higher dose ratios (CTDI 450 / CTDI 150 )for the bottom and center positions aredue to the higher scatter-to-primary ratios since the central dose peak does not change when comparing the 150 mm and 450 mm lengthscans. Figure 5-4 simultaneously displays the helical scan dose profiles (center position, head phantom, 0.5 pitch, 24 mm slice thickness) of the measured single phantom and the reflected 150 and 450 mm length phantoms. Figure 5-5similarly displays the three helical scan dose profiles measured in the top position of the PMMA head phantom for a pitch of 1.0 and 24 mm slice thickness. The figures provides strong visual evidence to the fact that the 150 mm length dose profile significantly fails to measure the scatter tails associated with longer scan lengths. Also evident is the fact that the center peak does not change as the phantom size is lengthened (i.e. accumulated dose differences in the corresponding profiles areattributed solely to scatter radiation). The simultaneous prediction of the helical dose profiles for both a 150 and 450 mm length phantom in a single measurement is especially helpful when working with heavy 32 cm diameter body phantoms. As explained for the head phantom, the FOC dosimeter was placed at z = 7.5 cm in the 300 mm length body phantom and the measured profile was reflected about the peak dose ( z = 7.5 cm) to predict the helical scan dose profiles for both the 150 and 450 mm phantom lengths. Accumulated dose ratios for the two predicted dose profiles in the 32 cm diameter body phantom are listed in Table 5-7. The higher dose ratios (CTDI 450 / CTDI 150 )for the bottom and center positions (compared to the other peripheries) are due to the higher scatter-to-primary ratios (SPRs)at these positions since the central dose peak does not change when comparing the 150 mm and 450 mm length scans. The bottom and center ratios are also higher when compared

PAGE 86

86 to head phantom data (Table 5-5) since the SPRs increase with increasing phantom diameter thickness. Figure 5-4. Helical scan dose profiles in the center position of CTDI head phantoms measured with fiber-optic coupled dosimeters. CT scans were performed at 120 kVp and 82 effective mAs for a selected pitch of 0.5 and a slice thickness of 24 mm. Plotted are the dose profiles measured in a single (15 cm length) head phantom, the reflected first 7.5 cm of a dose profile measured in a 300 mm length phantom with FOC dosimeter placed at z = 7.5 cm, and the reflected 450 mm length profile. Table 5-7 also displays a significant increase in dose ratios with corresponding increases in pitch. This trend is also seen in the head phantom ratios (Table 5-5, 5-6), but not as extensively as in the larger diameter body phantom. As pitch is increased, scatter-to-primary ratios increase as well at points further away from the peak dose. Thus, for dose profiles acquired in longer (450 mm) length phantoms that collect the extended scatter tails not measured in shorter (150 mm) phantoms, dose ratios increase with pitch. In the body phantom, where scatter-to-primary ratios are already high to begin with, this trend can be even more significant.

PAGE 87

87 Figure 5-5. Helical scan dose profiles in the 12 oclock (top)position of CTDI head phantoms measured with fiber-optic coupled dosimeters. CT scans were performed at 120 kVp and 82 effective mAs for a selected pitch of 1.0and a slice thickness of 24 mm. Plotted are the dose profiles measured in a single (15 cm length) head phantom, the reflected first 7.5 cmof a dose profile measured in a 300 mm length phantom with FOC dosimeter placed at z = 7.5 cm, andthe reflected 450 mm length profile. Table 5-7.Accumulated dose ratios measured from helical dose profiles of 450 mm and 150 length PMMA body phantoms. Single phantom (150 mm length) dose profiles were predicted from the reflected first 7.5 cm ( z = 0 to 7.5 cm) ofthe double phantom study, while z = 7.5 to 30 cm was reflected about the peak dose ( z = 7.5 cm) to predict the helical dose profiles for a 450 mmlength phantom. slice thickness 12 mm 24 mm pitch 0.5 1 1.5 0.5 1 1.5 module CTDI 450 / CTDI 150 ** Center 1.31 1.35 1.42 1.43 1.47 1.56 Top 1.10 1.13 1.24 1.27 1.38 1.58 Left 1.19 1.28 1.33 1.30 1.38 1.38 Bottom 1.33 1.47 1.52 1.43 1.49 1.58 **Calculated from reflected z = 0 to 7.5 cm for the 300 mm lengthphantom dose profile

PAGE 88

88 CTDI 100 Efficiency Revisited The 450 was determinedas the ratio of the accumulated dose integral measured in the center 100mm of the predicted head and body phantomdose profilesdivided by the total 450 mm scan length of the PMMA phantom that includes the scatter edges. Also listed are the 150 efficiencies measured above and the corresponding CTDI 100 efficiences simulated by Boone 25 Table 5-8. CTDI100 efficiency (CTDI100 / CTDIL ) calculated as the ratio of the accumulated dose measured in the center 100mm of a standard PMMAphantom divided by the total scan length L of thephantom. CTDI100 efficiencies were calculated for 150 mm ( 150 ) and 450 mm ( 450 ) length dose profiles. Also listed are Monte Carlo simulations performed by Boone for calculating CTDI100 efficiency ( ) as the ratio of the accumulated dose measured in the center 100mm of an infinitely long PMMA phantom divided by the total dose in the infinite phantom. Center Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm 150 0.92 0.95 0.90 0.95 450 0.71 0.71 0.59 0.61 0.82 a 0.81 b 0.63 a 0.62 b Peripheries avg Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm 150 0.92 0.95 0.92 0.95 450 0.74 0.74 0.72 0.70 0.90 a 0.89 b 0.88 a 0.87 b a Calculated with profiles acquired at slice thickness of 10 mm b slice thickness of 20 mm

PAGE 89

89 The 450 efficiency at the four peripheries (top, left, right, bottom) has been averaged and compared with corresponding values at the center position of the two phantoms. Profiles predicted at varying pitches (0.5, 1.0, 1.5) for a given slice thickness showed little change, given by the variance (Table 5-9). Therefore, the 450 efficiencies listed were calculated as the average of measured values at the three pitches for each given slice thickness and phantom position permutation. Table 5-9.Variance of 450 efficiencies calculated from helical scan dose profiles acquired at the three pitches for each combination of slice thickness and phantom position utilized. Scans were predicted for thefive positions of both the head and body PMMA phantoms for two slice thicknesses at each position. Head Phantom Body Phantom slice thickness 12 mm 24 mm 12 mm 24 mm Variance Center 1.39% 1.86% 0.50% 0.47% Peripheries avg 3.09% 3.35% 2.90% 7.26% The 150 efficiencies measured earlier showed goodagreement with Boones data at all slice thicknesses for peripheries of both the head and body phantoms. Such agreement was not seen, however, at the center of the two phantomsdue to the finite size of the 150 mm length phantom,resultingin the inability to record the scatter of larger beam widths that extend beyond the limits of the phantom. The 450 efficiencies, on the other hand, show good agreement with Boones simulations in the center position as thelonger length (450 mm) phantom is sufficient in collecting the scatter tail contributions at the edges of the dose profiles. The 450 however,

PAGE 90

90 overestimates the scatter contributions in the peripheries when compared to the other CTDI efficiency metrics ( 150 ). Conclusions Analyses of single scan dose profiles measured with standard pencil ion chambers 22,25-26 have illustrated the shortfall of traditional CTDI to accurately predict the dose delivered from helical image acquisitions. Fiber-optic dosimeters ability both discriminate the scatter tails associated with helical CT and measure dose for clinically relevant scan lengthswhere ion chamber collected charge measurements fail givescredence to the already established thought that longer phantoms are neededfor clinical CT dosimetry.47,49 Novel Method for Predicting Helical Scan Dose Profiles Outlined above isa novel and yet simple method for predicting helical scan dose profiles using fiber-optic coupled dosimeters in extended length phantoms. The active area of the fiberoptic dosimeter was placed at z = 7.5 cmin a 300 mm length PMMA phantom and CT acquisitions were performed for the entire length of the phantom ( L = 300 mm). When the dose profile is plotted versus phantom position, the dosepeak occurs at z = 7.5 cm (Figure 5-1). By reflecting both sides of the dose profile across the center line in the peak dose, helical dose profiles for both a 150 and 450 mm length phantomcan be simultaneously predictedin a single measurement. Need for Longer Phantoms Accumulateddose ratios (CTDI450 / CTDI 150 ) of these predicted profiles verify the failure of CTDI measurements, performed in a standard 150 mm length phantom, to sufficiently measure the scatter radiation of clinically relevant scan lengths. Figures 4-4 and 4-5 illustratethe fact that 150 mm length dose profiles fail significantly to measure the scatter tails associated with longer scan lengths. The figures show that the center peak does not change as the phantom

PAGE 91

91 size is lengthened andthus, accumulated dose differences in longer (e.g. 450 mm length) phantoms when compared to shorter (150 mm) phantom profiles are ascribed almost exclusively to scatter radiation.

PAGE 92

92 CHAPTER 6 CONCLUSIONS Prototypical Fiber-Optic Coupled Point Dosimetry System Fiber-optic coupled dosimeters based on the sensitive elements of acoupled scintillation phosphor demonstrate strong sensitivity, reproducibility, and excellent dose linearity across the diagnostic energy range. The dosimetry systems radiation detection capabilities are due to the absorption of x-rays in a gadolinium oxy-sulfate scintillation phosphor material that is coupled to the end of a fused silica, 400 m diameter, optical fiber. The phosphors absorption of incident x-rays via the photoelectric effect provides a conversion of x-ray photon energy to visible light, which then travels the length of the fiber to a photomultiplier tube (PMT). The PMT converts the incident light photons to an output voltage signal proportional to the number of light photons incident at the photomultiplier tube interface. This output is then relayed in terms of (photon) counts to a PC. The FOC point dosimeterapproach provides remotedetection of the dose associated with diagnostic imaging modalities such as x-ray and computed tomography (CT). Thereal-time response of the system allows for the direct recording of helical scan dose profiles that contain the essential information for the characterization of the dosimetry quantities fundamental to multi-detector computed tomography. The acquisition of empirical helical scan dose profiles using FOC dosimeters distinguishes the primary radiation from the scatter contributions that makeup the dose delivery of a given CT system and isolate the effect of specific scanner parameters on patient absorbed dose. Helical Computed Tomography DoseProfilesHelical scan dose profiles measured with fiber-optic coupled dosimeters and plotted versus time and/or phantom position provide information of CT delivery characteristics that more

PAGE 93

93 traditionalcomputed tomography dosimetry metrics lack. They portray the periodic motion of helical scanning from above to below the table top, clearly seenin the peaks and valleys of the measured dose profilesas the scanner cycles around the dosimeter. Peaks in the dose profile correspond to an x-ray tube position closest to the detector. Corresponding minima occur when the source-to-detector distance is at a maximumand demonstrate the effect ofphantom and/or table attenuation on patient absorbed dose. Dose profiles acquired in the center position of CTDI phantoms show both a lowering and flattening out of the central dose peak due to the symmetry of x-ray attenuation from the cylindrical phantom and the constant source-to-detector distance. The scatter tails not measured by conventional CTDI 100 are also seen in helical dose profilesthat scanthe entire length of the phantom. Helical scan dose profiles display the attenuation effects of PMMA material and the CT table itself when fiber-opticdosimeters are placed in the various positions of the CTDI phantom. Body phantom data shows an even more significant reduction of dose at the center of the scan length in the measured profile as a result of the increased amount of phantom material when a larger diameter phantom is chosen. The lowering and flattening of the central dose and the broadening of the scatter tails in the central position dose profile is also enhanced due to the increased x-ray attenuation and the higher scatter-to-primary ratio of the larger diameter body phantom. The measurement of single scan dose profiles only provides information regarding a single axial slice. Thus, the effect on changes in dose delivered due to a change in pitch in contiguous scanning cannot be determined from conventional SSDPs. However, helical scan dose profiles measured with fiber-optic coupled dosimeters can provide information regarding the characteristic effect of pitch on a CT systems dose delivery. Measured data shows the increased

PAGE 94

94 dose typical of overscanning when scanner pitch is less than 1.0, despite the scanners use of effective mAs to maintain dose with changes in pitch. The primary (central peak) dose delivered to a cylindrical PMMA phantom does not change significantly when pitch ischanged. Therefore, the overscanning typical of low pitch selection leads to an increase in dose due to FOC dosimeter efficient measurement of the scatter radiation at the edges of standard dosimetry phantom lengths. Novel Method for Predicting Helical Scan Dose Profiles Analyses of single scan dose profiles measured with standard pencil ion chambers have illustrated the shortfall of traditional CTDI to accurately predict the dose delivered from helical image acquisitions. Fiber-optic dosimeters abilityboth discriminate the scatter tails associated with helical CT and measure dose for clinically relevant scan lengthswhere ion chamber collected charge measurements fail gives credence to the already established thought that longer phantoms are needed forclinical CT dosimetry. A novel and yet simple method for predicting helical scan dose profiles using fiber-optic coupled dosimeters in extended length phantomswas developed in this study. The active area of the fiber-optic dosimeter was placed at z = 7.5 cm in a 300 mm length PMMA phantom and CT acquisitions were performed for the entire length of the phantom ( L = 300 mm). When the dose profile is plotted versus phantom position, the dose peak occurs at z = 7.5 cm. By reflecting both sides of the doseprofile across the center line in the peak dose, helical dose profiles for both a 150 and 450 mm length phantom can be simultaneously predicted in a single measurement. Need for Longer Phantoms Accumulated dose ratios (CTDI 450 / CTDI 150 ) of these predicted profiles verify the failure of CTDI measurements, performed in a standard 150 mm length phantom, to sufficiently measure the scatter radiation of clinically relevant scan lengths. An evaluation of the CTDI 100

PAGE 95

95 metric in various length phantoms along with plots of their corresponding dose profiles (Figures 4-4 and 4-5)show that the center peak does not change as the phantom size is lengthened. Thus, accumulated dose differences in longer (e.g. 450 mm length) phantoms when compared to shorter (150 mm) phantom profiles are ascribed almost exclusively to scatter radiation.Future Work Extended Characterization of the Point Dosimetry System Fiber-optic coupled dosimeters show strong energy dependence, specifically sensitivity to lower energies in the diagnostic range, due to the photoelectric absorption K-edge (50 keV) of the gadolinium oxy-sulfate scintillation phosphor. An extended characterization of the point dosimetry system is therefore needed to calibrate fiber response relative to the energy technique chosen. An inherent angular dependence seen in the fibers coupled active area must also be readdressed. The sinusoidal response with perpendicular-to-axial fiber positioning can be significantly reduced when the dose measurements are performed within a scattering medium. An angular dependence for axial positioning, when the x-ray tube is directed head-on still remains. Verification of Predicted Dose Profiles in Extended Length Phantoms This study provided a simple method for determining the helical scan dose profiles of extended length phantoms when extra phantom equipment is either unavailable or too cumbersome to transport. By placing the FOC dosimeter at z = 7.5 cm in a 300 mm length phantom, two helical scan dose profiles could be predicted (150 and 450 mm length phantoms) in a single measurement. The effectiveness of the 150 mm length phantom predicted dose profiles was verified when compared to measured helical scan dose profiles across the entire 150 mm length phantom. Central peak and scatter dose contributions to the profiles showed good correlation with integral dose measurements across the profile length having strong agreement.

PAGE 96

96 Physical measurements 450 mm length PMMA phantom profiles are therefore needed to verify the accuracy of the corresponding predicted profiles for this phantom length. Multi-Fiber-Optic Coupled Dosimeter System Development of a multi-fiber system to allow for the simultaneous recording of multiple absorbed dose measurements has begun in the Nuclear & Radiological Engineering Department at the University of Florida under the advisement of Dr. David Hintenlang. The fiber-optic coupled dosimeters are based on the same properties of the couple scintillation phosphors as described in this study. Multiple photo-multiplier tubes and a custom designed software program allow for the recording of simultaneous dose measurements. Anthropomorphic Phantoms These dose profiles at the center and peripheries of CTDI phantoms indicate the differences in dose delivery to organs within the body, and will be improved with applications with anthropomorphic phantom. Such aseries of anatomical physical phantoms have been fabricated for research applications at the University of Florida (UF). 51,52 A unique feature of this family of phantoms is that each one has been developed to correspond precisely with a complementary computational model, permitting a direct correlation of organ dosimetry between computational simulations and empirical measurements for identical models. The phantom series has been constructed to represent a range of ages and incorporates real-time dosimeters for a large number of organ locations in order to perform rapid measurements of both specific organ doses and Effective Dose. 53-57 The early phantoms demonstrated the value of a physical phantom that could be utilized for dosimetry measurements at a wide variety of clinical facilities, and were the first to incorporate an array of immediate read-out dosimeters that facilitated the examination of doses for a series of radiographic examinations and varying techniques. 58 Since the phantoms were

PAGE 97

97 also identically matched to a computational model, they also provided a unique opportunity to benchmark and compare the results of physical measurements with computational simulations. 59 More recent investigations have focused on doses delivered from helical and MDCT systems. Jones 56 performed a comprehensive evaluationof the effect of CT acquisition parameters on organ doses and Effective Dose of a newborn. Currently,a series of adult physical phantoms are under developmentto evaluate the absorbed dose associated with CT image acquisitions in both male and female patients. Final Thoughts Fiber-optic coupled dosimeters provide the potential for improvements in standard CT dosimetry where pencil ion chamber methods fail to sufficiently measure the scatter radiation associated with large diameter phantoms and longer, more clinically relevant scan lengths. Helical scan dose profiles measured with FOC dosimeters differentiate the primary radiation from the scatter contributions associated with a given CT system to evaluate the effect of specific scanner parameters on patient absorbed dose. The potentialto measure specific organ dose and effective dose measurementsusing fiber-opticdosimeters along with anatomical physical phantoms can allow a description of CT dose beyond the limits of standard dosimetr

PAGE 98

98 LIST OF REFERENCES 1. American Association of Physicists in Medicine, AAPM Report 96,The Measurement, Reporting, and Management of Radiation Dose in CT (AAPM, College Park, MD, 2008). 2. FDA,FDA public health notification,reducing radiation risk of computed tomography for pediatric and small adult patients, Pediatr.Radiol.32 314-216(2002). 3. E. L. Nickoloff, and P. O. Alderson,Radiation exposures to patients from CT.Reality, public perception, and policy, AJR Am.J.Radiol.177 285-287(2001). 4. H. Hu,Multi-slice helical CT,Scan and reconstruction,Med.Phys.26 5-18(1999). 5. C. H. McCollough, and F. E. Zink,Performance evaluation of a multi-slice CT system, Med.Phys.26 2223-2230(1999). 6. T. G. Flohr, K. Stierstorfer, S. Ulzheimer, H. Bruder, A. N. Primak, and C. H. McCollough,Image reconstruction and image quality evaluation for a 64-slice CT scanner with z-flying focal spot, Med.Phys.32 2536-2547(2005). 7. T. G. Flohr, S. Schaller, K. Stierstorfer, H. Bruder B. M. Ohnesorge, and U. J. Schoepf, Multi-detector row CT systems and image-reconstruction techniques, Radiology 235 756-773(2005). 8. S. Mori, M. Endo, T. Tsunoo, S. Kandatsu, S. Tanada, H. Aradate, Y. Saito, H. Miyazaki, K. Satoh, S. Matsushito, and M. Kusakabe,Physical performance evaluation of a 256slice CT-scanner for four-dimensional imaging, Med.Phys.31 ,1348-1356,(2004). 9. IMV,Benchmark Report CT,(In:Young L,ed.Des Plaines, IL,IMV Medical Information Division, Inc., 2006) www.IMVinfo.com 10. O.W. Linton, and F.A. Mettler,Jr., National conference on dose reduction in CT, with an emphasis on pediatric patients, AJR Am.J.Roentgenol.181 ,321-329,(2003). 11. F.A. Mettler, Jr., P. W. Wiest, J. A. Locken, and C. A. Kelsey,CT scanning:Patterns of use and dose, J.Radiol.Prot.20 ,353-359,(2000). 12. J. T. Bushberg, J. A. Seibert, E. M. Leidholdt, and J. M. Boone, The Essential Physics of Medical Imaging ,(Lippincott Williams & Wilkins,Philadelphia, Pennsylvania,2002). 13. International Atomic Energy Agency,IAEA Safety Series No.115,International Basic Safety Standards for Protection Against Ionizing Radiation and for the Safety of Radiation Sources(IAEA, Vienna, Austria,1996). 14. American Association of Physicists in Medicine, AAPMReport 74,Quality Control in Diagnostic Radiology,(AAPM, Chicago, IL,2002).

PAGE 99

99 15. U.S.FDA Code of Federal Regulations,Diagnostic x-ray systems and their major components,21 CFR1020.33,Govt. Printing Office, 1984. 16. American Association of Physicists in Medicine, AAPM Report 31,Standardized Methods for Measuring Diagnostic X-rayExposures, (AAPM,New York,1990). 17. American Association of Physicists in Medicine, AAPM Report 39,Specification and Acceptance Testing of Computed Tomography Scanners,(AAPM, NewYork,1993). 18. International Electrotechnical Commission,International Standard 61223-2-6, Evaluation and Routine Testing in Medical Imaging Department,Part 2-6:Constancy Tests -X-ray Equipment for Computed Tomography,(IEC,Geneva, Switzerland,1994). 19. American Association of Physicists in Medicine, AAPM Report 1,Phantoms for Performance Evaluation and QualityAssurance of CT Scanners,(AAPM, Chicago, IL, 1997). 20. K. A. Jessen, W. Panzer, P. C. Shrimpton, et al.,EUR 16262,European Guidelines on Quality Criteria for Radiological and Electromedical Healthcare IT Industry (COCIR), Frankfurt,2000 21. International Electrotechnical Commission,International Standard 60601-2-44, Edition 2.1, Medical Electrical Equipment-Part 2-44:Particular Requirements for the Safety of X-Ray Equipment for Computed Tomography(IEC,Geneva, Switzerland, 2001). 22. H. Y. Tsai, C. J. Tung, M. H. Huang, and Y. L. Wan,Analyses and applications of single scan dose profiles in computed tomography, Med.Phys.30 ,1982-1989(2003). 23. T. Cheung, Q. Cheng, D. Feng, M. J. Stokes,Study on examinees dose delivered in computed tomography, Phys.Med.Biol.46 ,813-820(2001). 24. T. B. Shope, R. M. Gagne, and G. C. Johnson,A method for describing the doses delivered by transmission x-ray computed tomography, Med.Phys.8 ,488-495(1981). 25. J. M. Boone,The trouble with CTDI 100, Med.Phys.34 ,1364-1371(2007). 26. R. L. Dixon,A new look at CT dose measurement,Beyond CTDI, Med.Phys.30 1272-1280(2003). 27. J. M. Boone, V. N. Cooper III, W. R. Nemzek, J. P. McGahan, J. A. Seibert,Monte Carlo assessment of computed tomography dose to tissue adjacent to the scanned volume, Med.Phys.27 ,2393-2407(2000). 28. J. Hecht, Understanding Fiber Optics ,(Pearson Education, Inc., Upper Saddle River, NJ, 2006).

PAGE 100

100 29. A. L. Huston, B. L. Justus, P. L. Falkenstein, R. W. Miller, H. Ning, and R. Altemus, Remote optical fiber dosimetry, Nucl.Instrum.Methods Phys.Res.B 184 ,55 (2001). 30. A. L. Huston, B. L. Justus, P. L. Falkenstein, R. W. Miller, H. Ning, and R. Altemus, Optically stimulated luminescent glass optical fiber dosemeter, Radiat.Prot.Dosim.101 ,23(2002). 31. B.L. Justus, C. D. Merritt, K. J. Pawlovich, A. L. Huston, and S. Rychnovsky,Optically stimulated luminescence dosimetry using doped fused quartz, Radiat.Prot.Dosim.84 189(1999). 32. B. L. Justus,P.L. Falkenstein, A. L. Huston, M. C. Plazas, H. Ning, and R. W. Miller, Gated fiber-optic-coupled detector for in vivo real-time radiation dosimetry, Optical Society of America 43 (8),1663-1668(2004). 33. M. J. Marrone,Appl.Phys.Lett.38 (3) 115(1981). 34. www.oceanoptics.com 35. R. D. Evans, The Atomic Nucleus ,(Krieger Publishing Company, Inc., Malabar, FL, 1982). 36. K. Ursel, and A. Richards A,Kodak Publication No.M3-103, Kodak MIN-R 2000 Mammography Screen/Film Systems User Guide,(Eastman Kodak Company,2006). 37. W. E. Moore, R. Dickerson, and D. Steklenski, Design and performance features of a new mammographic film/screen system, No. 54, Proc. SPIE, 5368 (2004). 38. www.e-radiography.net/radtech/f/film.htm 39. HamamatsuPhotonics K.K., Hamamatsu Preliminary Data: Photon Counting Head with Microcontroller and RS-232C Interface, H7467 Series, (Nov. 1999). www.hamamatsu.com 40. G. F. Knoll, Radiation Detection and Measurement (John Wiley & Sons,Inc., New York, 2000). 41. M. Mahesh, J. C. Scatarige, J. Cooper, and E. K. Fishman,Dose and pitch relationship for a particular multi-slice CT scanner, AJR Am.J.Roentgenol.177 ,1273-1275(2001). 42. MHRA Evaluation Report MDA 04037,ImPACT Report:Siemens SOMATOM Sensation 16 CT Scanner,(MHRA,London, 2004). 43. W. Huda,Dose and image quality in CT, Pediatr.Radiol.32 ,709-713(2002). 44. D. J. Brenner, E. J. Hall, and D. Phil,Computed tomography an increasing source of radiation exposure, N.Eng.J.Med.357 ,2277-2284(2007).

PAGE 101

101 45. D. J. Brenner DJ,Is it time to retire the CTDI for CT quality assurance and dose optimization? Med.Phys.32 3225-3226(2005). 46. D. J. Brenner, C. H. McCollough, and C. G. Orton,Is it time to retire the computed tomography dose index (CTDI) for CT quality assurance and dose optimization, Med. Phys. 33 ,1189-1191(2006). 47. K. D. Nakonechny, B. G. Fallone, and S. Rathee,Novel methods of measuring single scan dose profiles and cumulative dose in CT, Med.Phys.32 ,98-109(2005). 48. R. L. Dixon, M. T. Munley, and E. Bayram,An improved analytical model for CT dose simulation with a new look at the theory of CT dose, Med.Phys.32 ,3712-3728(2005). 49. R. L. Dixon, and A. C. Ballard,Experimental validation of a versatile system of CT dosimetry using a conventional ion chamber:Beyond CTDI100 Med.Phys.34 (8),33993413(2007). 50. H. Zhou, and J. M. Boone,Monte Carlo evaluation of CTDI in infinitely long cylinders of water, polyethylene and PMMA with diameters from 10 mm to 500 mm, Med.Phys.35 (6),2424-2431(2008). 51. M. A. Tressler, D. E. Hintenlang,"Construction of a newborn dosimetry phantom for measurement of effective dose," Health Phys. 76 (6) S190(1999). 52. A. K. Jones, T. A. Simon, W. E. Bolch, M. M. Holman, and D. E. Hintenlang,A tomographic physical phantom of the newborn patient with real-time dosimetry I: Methods and techniques for construction, Med.Phys.33 (9) 3274-3282(2006). 53. M. Bower, and D. E. Hintenlang,The characterization of a commercial MOSFET dosimeter system for use in diagnostic x-ray, Health Phys.75 ,197-204(1998). 54. A. K. Jones, F. D. Pazik, D. E. Hintenlang, and W. E. Bolch,MOSFET dosimeter depthdose measurements in heterogeneous tissue-equivalent phantoms at diagnostic x-ray energies, Med.Phys.32 (10),3209-3213(2005). 55. R. J. Staton,A. K. Jones, C. Lee, D. E. Hintenlang, M. M. Arreol, J. L. Williams, and W. E. Bolch,A tomographic physical phantom of the newborn child with real-time dosimetry II:Scaling factors for calculation of mean organ dose in pediatric radiography, Med.Phys.33 (9),3283-3289(2006). 56. A. K. Jones,Dose versus image quality in pediatric radiology,Studies using a tomographic newborn physical phantom with an incorporated dosimetry system, Ph.D. Dissertation, University of Florida,Gainesville, FL(2006). 57. A. K. Jones, A. L. Huston, P. L. Falkenstein, and D. E. Hintenlang,Evaluation of a fiber optic coupled dosimeter for use in computed tomography dose measurements, Radiat. Prot.Dosimetry; In Review(2008).

PAGE 102

102 58. K. M. Hintenlang,J. L. Williams, and D. E. HintenlangDE,A survey of radiation dose associated with pediatric diagnostic x-ray, Pediatr.Radiol.32 ,771-777(2002). 59. J. B. Sessions,J. N. Roshau,M. A. Tressler, D. E. Hintenlang, M. M. Arreola, J. L. Williams, L. G. Bouchet,and W. E. BolchWE,Organ doses in pediatric radiology:A comparison of an anthropomorphic physical phantom with MOSFET dosimetry and a computational patient model, Med.Phys.29 (6)(2002).

PAGE 103

103 BIOGRAPHICAL SKETCH William Edward Moloney was born in Brooklyn, New York, on May 1, 1982, to William and Jane Moloney. He is one of five children including his older sister Liz and three younger brothers Matthew, John, and Robert. He attended Xaverian High School in Brooklyn, New York, graduating in 2000. Hethenearned his Bachelor of Science degree in physics in 2004 from Saint Josephs University in Philadelphia, Pennsylvania. In August 2008, he received his Master of Science degree in nuclear engineering sciences with a concentration in medical physics fromthe University of Florida in Gainesville, Florida.