Citation
A Computer-Based Simulation for Training Dental Preparation

Material Information

Title:
A Computer-Based Simulation for Training Dental Preparation
Creator:
Hunt, William C
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (41 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Computer Engineering
Computer and Information Science and Engineering
Committee Chair:
Peters, Jorg
Committee Members:
Lok, Benjamin C
Fishwick, Paul A
Graduation Date:
8/11/2012

Subjects

Subjects / Keywords:
College students ( jstor )
Coordinate systems ( jstor )
Data models ( jstor )
Dental models ( jstor )
Drilling ( jstor )
Maps ( jstor )
Simulations ( jstor )
Teeth ( jstor )
Tooth enamel ( jstor )
Virtual reality ( jstor )
Computer and Information Science and Engineering -- Dissertations, Academic -- UF
dental -- drilling -- graphics -- haptic -- preparation -- simulation -- training -- virtual
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Computer Engineering thesis, M.S.

Notes

Abstract:
We present a low-cost VR simulation environment for dental students that allows trainees to practice dental preparation. The trainee haptically interacts with a virtual tooth, drilling into its enamel and dentin layers. When training students, the environment can compute metrics, saving educators’ time. The system was tested and evaluated by dental students and professors of dentistry. ( en )
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.
Thesis:
Thesis (M.S.)--University of Florida, 2012.
Local:
Adviser: Peters, Jorg.
Statement of Responsibility:
by William C Hunt.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Hunt, William C. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2012 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

ACOMPUTER-BASEDSIMULATIONFORTRAININGDENTALPREPARATIONByWILLIAMCHRISTOPHERHUNTATHESISPRESENTEDTOTHEGRADUATESCHOOLOFTHEUNIVERSITYOFFLORIDAINPARTIALFULFILLMENTOFTHEREQUIREMENTSFORTHEDEGREEOFMASTEROFSCIENCEUNIVERSITYOFFLORIDA2012

PAGE 2

c2012WilliamChristopherHunt 2

PAGE 3

Idedicatethisworktomymother:Wellyouneedtoatleastgetamaster's...It'swhatabachelor'smeantwhenIwasyourage! 3

PAGE 4

ACKNOWLEDGMENTS Dr.JorgPetershasbeentheidealthesissupervisor.Hisfrankcriticism,condence,andpatientencouragementaidedthecreationofthisprojectandthewritingofthisthesisininnumerableways. 4

PAGE 5

TABLEOFCONTENTS page ACKNOWLEDGMENTS .................................. 4 LISTOFTABLES ...................................... 6 LISTOFFIGURES ..................................... 7 ABSTRACT ......................................... 8 CHAPTER 1INTRODUCTION ................................... 9 1.1StatementofProblem ............................. 9 1.2EarlyWorkatUFSurfLab ........................... 10 1.3DenitionofTerms ............................... 10 2RELATEDWORK .................................. 12 3SYSTEMIMPLEMENTATION ............................ 14 3.1Overview .................................... 14 3.2DataModel ................................... 15 3.3Initialization ................................... 17 3.4ViewandControl ................................ 19 3.5Conguration .................................. 22 3.6Haptics ..................................... 22 3.7Sound ...................................... 26 3.8Ergonomics ................................... 27 4TESTINGANDEVALUATION ............................ 31 5FUTUREWORK ................................... 32 APPENDIX AUSERTESTINGINSTRUCTIONS ......................... 35 A.1Introduction ................................... 35 A.2Controls ..................................... 35 BSURVEYRESULTS ................................. 37 REFERENCES ....................................... 39 BIOGRAPHICALSKETCH ................................ 41 5

PAGE 6

LISTOFTABLES Table page 2-1Existingacademicdentalsystems ......................... 13 2-2Existingcommercialdentalsystems ........................ 13 B-1Quantitativesurveyresponses ........................... 38 6

PAGE 7

LISTOFFIGURES Figure page 3-1Basichigh-levelCEDSUMLClassDiagramshowingstateandbehavior .... 14 3-2Overcomingdisconnectedheightmapsandsteepnormalaliasingontooth36 18 3-3Examplesofgeneratedcarriesonteeth26-28and35-38 ............ 18 3-4CEDSdisplayloopSequenceDiagram ....................... 19 3-5Jaggededgesontooth36(FDInumbering) .................... 20 3-6Demonstrationofhandpieceorientationfeatures ................. 21 3-7GLConsoleInterface ................................. 22 3-8CEDShapticloopSequenceDiagram ....................... 23 3-9Demonstrationofdrilling ............................... 26 3-10CEDShapticloopActivityDiagram ......................... 27 3-11AnARmarkerattachedtoaheadband ....................... 28 3-12AdentalhandpieceattachedtoaNovintFalconRgrip .............. 28 3-13TheNovintFalconRwithattachedhandpieceorientedvertically ......... 29 3-14StudentsattheUFCollegeofDentistrytryingCEDS ............... 30 5-1Anaglyph3Dred/blueglasseswiththeCEDS'UF'ARMarkerAttached .... 32 7

PAGE 8

AbstractofThesisPresentedtotheGraduateSchooloftheUniversityofFloridainPartialFulllmentoftheRequirementsfortheDegreeofMasterofScienceACOMPUTER-BASEDSIMULATIONFORTRAININGDENTALPREPARATIONByWilliamChristopherHuntAugust2012Chair:JorgPetersMajor:ComputerEngineeringWepresentalow-costVRsimulationenvironmentfordentalstudentsthatallowstraineestopracticedentalpreparation.Thetraineehapticallyinteractswithavirtualtooth,drillingintoitsenamelanddentinlayers.Whentrainingstudents,theenvironmentcancomputemetrics,savingeducators'time.Thesystemwastestedandevaluatedbydentalstudentsandprofessorsofdentistry. 8

PAGE 9

CHAPTER1INTRODUCTION 1.1StatementofProblemTraditionaldentaltrainingmethodsarecostly,bothinmaterialandtime.Forexampledentalstudentsdestroytheirplasticpracticeteethordrillingpracticepads,somaterialmustconstantlybepurchased.Additionallydentalinstructorsmustspendtimetoindividuallyevaluatestudents'work.StudentsenteringintotheUniversityofFloridadentalprogramareintroducedtoworkingwithadentalhandpiecebydrillingonapracticepadcalledtheLearn-A-PrepII[ 1 ],butthishasbeenshowntonotsignicantlyimprovethelikelihoodofreceivinganAorBontheClassIIpractical(p=0.53)orontheComplexpractical(p=0.37)[ 2 ].Thisformoftrainingservestoacclimatedentalstudentstothedrill,butdoesnotfeelexactlyliketherealthing.Thesurfaceofthepracticepadshowscommontoothdecayshapesthatcanbetraced,anddifferenttoothlayersaremodeledwithdifferentcolors.Typodontsandplastictrainingteethhaveotherdownsides.Thesurfaceofadentoformisnotashardasarealtooth'senamelsurface.Furthermoretrainingteethareusuallyhomogeneous,withnosimulationofthevarioustoothlayers.Whenmovingthrougharealtooth,thedrillpassesthroughlayersofdifferinghardness(decay,enamel,dentin,andpulp).Beingfamiliarwiththefeelofmovingthroughthesedifferentmaterialtypesisnecessarytoavoidmakingmistakesonrealpatients,butplasticteethfailtopreparestudentsfully.Acomputer-basedreplacementcouldbereusableandlow-cost.Simulationsoftwarehasnoinheritlimitonthenumberofattemptsthatcanbemadeataremovalprocedure.Thistranslatesintoahighsavingsinmaterialcostsovertime.Additionallyaninstructorwouldonlyneedtosetthemetricsthatthestudents'workisgradedagainst.Thissavesteachersfromrepeatingthesamegenericevaluations,leavingtimeformore 9

PAGE 10

specializedone-on-oneattention.Asabonus,evaluationwouldbeinstant,accurate,andobjective.ThesesentimentsarereectedbyUFDentalProfessorKarl-JohanSoderholm: Thedevelopmentofaninexpensivedentalsimulator,capableofassistingdentalstudentsindevelopingandimprovingtheirpsychomotorskilllevelisverychallengingbutalsoveryworthwhile.Today,dentalstudentshavetospendmonthsinalaboratoryenvironmenttodevelopthoseskilllevels,atrainingthatisbothteacherdemandingaswellasexpensive.Byuseofasoftwarebaseddentalsimulator,thestudentswouldbeabletopracticeatanyplacetheysopreferandreceivefeedbackdirectlyfromthecomputer.Suchatrainingwouldbemuchmoreefcientandinexpensivethanthetrainingthatisofferedinthelaboratory. 1.2EarlyWorkatUFSurfLabTheUFSurabhasahistoryworkingwithmedicalapplications.TheToolkitforIllustrationofProceduresinSurgery(TIPS)isacollaborativepartnershipofsurgeonsandITresearchersleveraginglow-costhardwareandadvancedsoftware.Itsgoaliscreatingandbroadlydisseminatingalow-cost,computer-based,3Dinteractivemultimediaauthoringandlearningenvironmentincludingforcefeedbackforcommunicationofsurgicalprocedures.TheComputerEnhancedDentalSimulation(CEDS)projectfollowsinthesefootsteps.OuroriginalimplementationofCEDSwasfocusedonemulatingandultimatelyreplacingtheLearn-A-PrepIIpracticedevice[ 1 ].Thisversiondidnotuseanyforce-feedbackdevicesandwasinsteadcontrolledusingadrawingtablet.Studentswouldpracticetracingtheshapesfromthesurfaceofthepracticedrillingpad. 1.3DenitionofTerms ARToolkit:AsoftwarelibraryforbuildingAugmentedReality(AR)applications.Markersareidentiedinlivevideoandthepositionandorientationofthosemarkersareusedtosuperimposevirtualobjects[ 3 ].InCEDS,ARToolKitmarkers 10

PAGE 11

areusedtocreateaVirtualRealityeffectbyallowinguserstonavigatethe3Dsceneasthoughtheywerereallythere. Burr:Thesmallcutterattachedtothehandpiecethatcanremovetoothmaterialwhenrotatedathighspeeds.Dentalburrsaregenerallycylindricalwithvaryingbevel. Caries:Aprogressivedecayofbonestructure.Specicallyindentistry,ademineralizationofhardtissueanddestructionoforganicmatter. Dentin:Oneofthemajorcomponentsofatooth.Itisyellowandfoundjustbelowthetopenamel.Dentinislesshard,butalsolessbrittlethantheenamelitsupports. Dentoform:Theleadingdentalsimulationbrand,synonymouswitharticialsimulationteeth[ 4 ].WorkingonDentoformsimulationteethisgenerallythelasttrainingstepbeforebeginningworkonrealpatients. Enamel:Thehard,brittle,whitishtoplayeroftoothmaterial. FDIWorldDentalFederationNotation:NotationdevelopedbytheFDIWorldDentalFederationusedinternationallybydentiststoaspecifyteeth(ISO3950). Handpiece:Thepen-likedrillheldbythedentist.Burrscanbeswitchedoutonthehandpiece.Highspeedhandpiecescanspinashighas800,000RPMwhilelowspeedhandpiecesoperateatupto40,000RPMbutprovidehighertorque. Learn-A-PrepII:Adentaltrainingdeviceusedtofamiliarizedentalstudentswithoperatingthehandpiece.Itisarectangularpadwithshapesonthesurfacetotraceanddifferentcolorsatdifferentdepthstorepresentdifferentlayersofmaterial[ 1 ]. Preparation:Thedrillingnecessarytoremoveunwantedmaterialandinsertrestorativelling.Preparationsarebrokendowninto6preparationclasses.Whenadentalstudentpracticesapreparationonadentoformtooth,theyaresaidtobeperforminganidealpreparation. Pulp:Thesoftorganicmaterialfoundinsidethepulpchamberlocatedbelowthedentin. Typodont:Amodelmouthwhichholdsplasticteethfortraining. 11

PAGE 12

CHAPTER2RELATEDWORKSeveralattemptshavebeenmadetocreatehapticdentaltrainingsystems,butmostprojectsrelyonexpensivePhantomdevicesorspecialized3Dstereoscopicscreens.Furthermore,outoftheprojectscoveredbelow,onlyThomasetal.havetestedtheirsimulationandprovidedresults.Additionally,nonehavediscussedtheergonomicsoftheirsimulationsetup,nortechniquestoautomaticallyevaluatethedrilledtooth[ 5 ].Kimetal.proposeanextensivesetupwithstereoscopic3D,buttheirtoothmodelsappearratherroundedandunrealistic[ 6 ].Noborioetal.attempttoimproveonKimetal.witharichuserinterfaceandvariousburrshapes[ 7 ].Rhienmoraetal.provideamethodforrecordingandreplayinghapticmotionfortraining,buttheyonlymodifythedisplaymeshitselfwithoutalteringthenumberoftriangles(limitingthescaleofworkthatcanbedoneinapreparation)[ 8 ].Liuetal.providedetailedanalysisofrawforcesinvolvedindentalpreparation,butadmittohapticinstabilitywithrandomoscillations.Liuetal.alsoachievematerialremovalbyusingadeformablemesh[ 9 ].Wuetal.rendersurfacevoxelsdirectly,requiringbilinearinterpolationovernormalsforsmoothappearance,howeverevenwithnormallteringtheshapeoftheunderlyingvoxelstructureisvisible[ 10 ].Yauetal.runthemarchingcubesalgorithmonanadaptiveoctree-baseddatamodelthatcanhandlenedetail,butcanbecomplicatedtoupdate[ 11 ].Williametal.describearobustsurfacedraggingapproachthatweimplement,buttheyencounterablockyfeelingwhenworkingdirectlywiththevoxeldatastructure,whichwehavemanagedtomostlyavoid[ 12 ].FinallyThomasetal.onlyvisualizea2Dcross-section[ 5 ].Severalcommercialdentaltrainingsystemsexist,butmanyofthemareeitherstillindevelopmentordonotprovidedetailedtechnicalinformation. 12

PAGE 13

Table2-1. Existingacademicdentalsystems AuthorOverview Noborioetal.[ 7 ]Phantomdevice;QuickhullandGilbert,JohnsonandKeerthi(GJK)algorithmsonvoxelmodelKimetal.[ 6 ]Phantomdevice;3Ddisplay;AdaptivepolygonizationmethodandMauch'sclosestpointtransform(CPT)algorithmonvoxelmodelRhienmoraetal.[ 8 ]Phantomdevice;3Ddisplay;Forcesfromsurfacenormals;Deformablepolygonmesh;Hapticrecording,playbackfortrainingLiuetal.[ 9 ]Phantomdevice;Deformablepolygonmesh;HapticinstabilityWuetal.[ 10 ]Phantomdevice;Half-silveredmirror;Directrenderingofboundaryvoxels;FilteringofsurfacenormalsYauetal.[ 11 ]Phantomdevice;3Ddisplay;MarchingcubesonadaptiveoctreevoxelmodelWilliametal.[ 12 ]NovintFalconRdevice;Ballpivotalgorithm(BPA)onvoxelmodelThomasetal.[ 5 ]Impulsedevice;2Dcross-sectiondisplay Table2-2. Existingcommercialdentalsystems ProjectOverview VirtualRealityDentalTrainingSystem(VRDTS)[ 13 ]NovintFalconRSensableDental[ 14 ]PhantomVOXEL-MANDental[ 15 ]Phantomdevices;3DdisplayMoogSimodontDentalTrainer[ 16 ]OtherDentSim[ 17 ]Other 13

PAGE 14

CHAPTER3SYSTEMIMPLEMENTATION 3.1OverviewCEDSisanapplicationwritteninC++.Thesimulationisbrokenintomodulesrepresentingthedatamodel,initializer,viewandcontrol,dynamicconguration,haptics,andsound.Themodulescommunicatewitheachotherviaacentralsimulationobjectwhichpermitsaccesstothemodules'variousinterfacesusingthesingletonpattern. Figure3-1. Basichigh-levelCEDSUMLClassDiagramshowingstateandbehavior Twothreadsrunsimultaneouslyduringthesimulation.Thedisplaythreadchecksforanynewdrillingsincethelastdisplayloopexecutionandupdatesthesurfacenormalsforviewingifnecessary.ThedisplaythreadalsochecksforvisibleAugmentedReality(AR)markerstoorienttheviewingangle.Displayupdatesmustbeefcientenoughnottosignicantlyimpedetheframe-rate.ThedisplayloopissummarizedinFigure 3-4 14

PAGE 15

Thehapticthreadprovidesforcefeedback,playssoundifnecessary,andupdatesthedatamodelwhenmaterialshouldberemoved.Thehapticthreadmustexecutewithin1mstoavoidachoppyfeeling.ThisloopisdescribedinFigure 3-8 .Figure 3-1 isaUMLClassDiagramdescribingatahigh-levelthevariousencapsulatedmodulesintheCEDSsimulation.Thefollowingsectionsdescribeeachmoduleingreaterdetail.Wealsodiscusstheergonomicsofourphysicalsetup. 3.2DataModelTheCEDSdatamodelextendstheoriginaltablet-basedCEDSimplementationinwhichonlydownwarddrillingwaspossible.BecausetheNovintFalconRisrestrictedto3degreesoffreedom,andbecauseaclass-1dentalpreparationischaracterizedmainlybyrmverticaldrillingandlateralsweepingratherthanrotation,thetoothdatarepresentationisasimplepiecewise-linearheightfunctionthatdoesnotsupportundercutting.Thenishingtouchesofareal-life,idealpreparationinvolvesomeslightundercutting(toensurethatllingmaterialstayslockedinplace),butthebulkofbasicmaterialremovaloccursthroughtranslationofthehandpiece,notrotation.TheNovintFalconR'srestrictionsmayhelpnewerstudentsunderstandhowtormlyholdthehandpiecesteadyandproperlycutintoatooth.AstudyatOttawaUniversitysuggestedthatutilizingatoothpreparationsupportsystemthatlimitedastudent'sdegreesoffreedominthismannerwasusefultoachieveagreatercompetencyintoothpreparation[ 18 ].CEDSemploysasimpleheightmapapproachratherthanavoxelapproachtorepresentthetoothtopography.Weareabletorepresenttheentiretoothbyusingtwoheightmaps.Oneheightfunctionmodelsthetopofthetooth,andanothermodelsthebottom.Thissimpledesignusesminimalspaceandenablesdirectvisualizationwithhighsurfaceresolutionandaccuratesurfacenormalsonthetopofthetooth.Theheightmaphowever,doesnotenforcethesemanticsofthehapticinteraction.Wefoundaheightmapofsize150x150tobedetailedenoughtocreatearealisticrenderingand 15

PAGE 16

abelievablefeeling,whilesmallenoughforinteractivework.Datastructureresolutioncouldberaisedtosupportevenmoredetailedhapticinteraction,butwechosetokeepthesizesmallenoughtoeasilyrepresenttheentiretoothsurfaceinasingle2Doatarraywithoutrequiringanyextraoptimizations.Increasingtheresolutionoftraditional3Dvoxelrepresentationscanquicklybecometoocomputationallylimitingincomparison[ 12 ].Theheightmapdatastructurestoressurfacenormalsfordisplay.Itspublicobjectinterfaceprovidestwomethodsforworkingwithnormals.First,apointonthesurfacemaybespeciedasafuturenormalupdatepoint,andsecond,acommandmaybesenttorecalculatethenormalssurroundingthemostrecentupdatepoint.Whenthetoothsurfacebecomessufcientlysteepaliasingcanoccurandtheresolutioncanbecomevisuallyapparentontoothedges(wherethetopographyofthetoothappearsvertical).Onafreshlyinitializedtoothmodelthisaliasingisproblematicneartheedgesoftheheightmaps(onthesidesofthetooth).Weareabletoxthesenormalsininitializationwithoutmodifyingthealiasedunderlyinggeometry.Thisnormalmodicationprocesswillbeexplainedintheinitializationsection.Whendrillingreducesatopheighttobelowerthanacorrespondingbottomheight,theheightatthatcoordinateisremovedfromboththetopandbottomfunctions,andalladjacentheightmappointsaresettotheaverageoftheircorrespondingtopandbottomheights.Thisprocesssewstogethertheopeninginthetoothgeometrycreatedwhenaheightmappointisremovedfromdisplay.Themomentafteraheightmapcoordinateisremoved,butbeforeitssurroundingedgeissewntogether,istheonlytimewhenthemodelexistsinaninconsistentstateforrendering,butinpracticethisstephappenssoquicklyitcannotbeobserved.Finallythenormalscorrespondingtothenewedgesonthetopandbottomheightmapsareaveragedforsmootherappearance.Afterthedatastructureisinitialized,thetopheightmapissavedtolaterdeterminedepth.Whendepthsareaddedtogethertheyrepresenttotalmaterialremoved.Decay 16

PAGE 17

isstoredbydepth,extendingdownfromtheoriginaltoothheight.Afterdrillinghasoccurred,thepercentofdecayremainingcanbedeterminedeasilybycomparingdepthremovedwithdecaydepth.Similarly,theamountremovedofanyothertoothmaterialcanbecomputed.Thetoothisseparatedintolayersofincreasinglysoftmaterial:enamel,dentin,andpulp.Wecurrentlymodelthebordersbetweentoothlayersbydepthfromtheoriginalheightofthetoothanddistancelaterallyfromtheedgeofthetooth.Inrealitythedistancefromthetopofthetoothintothesubsequentlayersisnotasuniformandwouldideallybeconguredbya3Dmodelprovidedbyadentalexpert. 3.3InitializationTheinitializationmoduleparsesstandardOBJles.TheOBJmodelsusedinCEDSwerecreatedfromSTLlesrepresenting3Dscansofrealteeth.Thegeometryoftheoriginal3Dtoothscansweremanuallymodiedtobecompletelyenclosed,anecessaryconditiontofullyinitializetheheightmaps.Toinitializethetopandbottomheighmaps,eachtriangleintheinputtoothmodelisexamined,andheightmapcoordinatesarecomparedtooverlappingtriangles.Thetopheightmapmaintainsthemaximumheightcorrespondingtoanytrianglesatthatparticularcoordinate,whilethebottomheightmapmaintainstheminimumtriangleheights.Simplyinitializingthedatamodelcoordinateswithmaximumandminimumtriangleheightsisinsufcienttocreateacompletetoothrepresentation,becausetheedgesofthetopandbottomfunctionsarenotguaranteedtomeet.Afterinitializingthedatastructuretheedgepointsareidentiedandsewntogetherbyaveragingedgepoints.Specialcaremustbetakentocreatepropernormalsattheedges.Thetopandbottomnormalsatsuchpointsareaveraged.Aftersculptingthetoothsurface,wexaliasedsurfacenormalsneartheedges.Foreachdegenerateedgepointontheheightmapforwhichanormalcannotbe 17

PAGE 18

directlycomputed,wesearchradiallyoutwardforanearbynon-steepheightmappoint.Weborrowthexandycomponentsofthenearbynormalandcombineitwiththezcomponentoftheoriginalaliasednormalandrenormalizeittocreateaconvincingmodiednormal. ABeforesewing BInherentedgealiasing CModiededgenormalsFigure3-2. Overcomingdisconnectedheightmapsandsteepnormalaliasingontooth36(FDInumbering) Weprogrammaticallyinitializedecayusingadepth-rstsearchforlowpointsonthetopofthetoothstartingfromthedatamodelcenter.Thedecaywillllupvalleysonthetoothsurface. ATooth26 BTooth27 CTooth28 DTooth35 ETooth36 FTooth37 GTooth38Figure3-3. Examplesofgeneratedcarriesonteeth26-28and35-38(FDInumbering) 18

PAGE 19

3.4ViewandControlCEDSusesFreeGLUTtodrivetheOpenGLdisplayloopandcollectkeyboardandmouseinput. Figure3-4. CEDSdisplayloopSequenceDiagram Thedisplaylooptraversesthetopandbottomtoothheightmapsoncetodirectlyrenderthem.Foranylocationonthedatamodelwithatopheightassociatedwithit,abottomheightmustalsoexist.Theheightmapsarerenderedbydrawingeverypossiblyquadbetweenexistingheightmappoints.Foranycoordinatewithoutaheightvalue,noadjacentquadswillbedrawn.Thisdirectrepresentationoftheheightmapcreatesajaggededgegeometry,butthisedgecanonlybenoticedwhenviewingdirectlyfromabove,orfromveryclose.Figure 3-5 showsanexampleofthisjaggededge(alongthebottomofthetoptwoteeth),andhowitquicklybecomesunapparentwhennotviewedfromveryclose.CEDSrequiresnospecialthreadsynchronizationfordisplay.Kimetal.pointoutthatmanyexistingsculptingsystemsrelyonuniformpolygonizationslikethemarchingcubesalgorithm,whichcanbecomputationallylimiting[ 6 ].Thesceneisviewedusingavirtualcameramountedonasphericalcoordinatesystemwiththetoothtopatthespherecenter.Thedefaultcameraorientationassumesthatthehandpieceisbeingmanipulatedbytherighthand,howeveraleft-handeduser 19

PAGE 20

Figure3-5. Jaggededgesontooth36(FDInumbering) canchoosetorotatethescene180degreestocompensateforthis.AstudentcannavigatethesceneusinganARToolKitmarkerattachedtotheirhead,givingaVirtualRealityeffectwhentheviewermovesthemarkerrelativetothecamera[ 3 ].Drawingthehandpieceonthetoothsurfaceratherthanatthevirtualdevicetipisimportanttomakethesurfaceappearhardtotheuser.Whenthevirtualhandpieceisdraggingonthetoothsurfacethephysicaldevicelocationdipsintothevirtualtooth.Ifthehandpiecemodelwererenderedatthisactualdevicelocationitwouldappeartobesinkingintothetoothsurface.Seeingthispenetrationwouldbetraytheuser'snotionofhowametallichandpieceshouldinteractwithahardtooth.Toavoidthisproblemthevirtualhandpieceisrenderedatwhatwecallthehapticanchorpoint,whichwillbeexplainedthoroughlyinthehapticsectionbelow.Thisshouldbethepointwheretheuserwouldexpectaphysicalhandpiecetoappear. 20

PAGE 21

Colorsfordecayandtoothlayersarewrittentoasingle3Dtextureafterthetoothisloaded.Decaydepthsarereadfromtheinitializedheightmap,andtoothlayersarederivedusingaconstantdepthfromthetoothsurface. AHandpieceinfrontoftooth BHandpieceabovetooth CHandpiecebehindtooth DHandpiecewitharrowFigure3-6. Demonstrationofhandpieceorientationfeatures Initiallyitcanbedifculttopositionthevirtualhandpieceonthevirtualtooth.Humansrelyonstereoscopicvisiontonavigatereal3Dspace.Wehaveusedsometrickstohelptheuserorientthevirtualhandpiecerelativetothevirtualtooth.Firstweprojectatransparentlinedownverticallyfromtheburr.Basedonhowthislineintersects,obscures,orisobscuredbythetooth,theusercanintuitivelyplacethevirtualburronthetoothsurface.Thesecondorientationfeatureisared,transparentarrowthatpointstothehandpiecefromtheburr.Ifthehandpieceleavesthescreentheusercaneasilynditandbringitbackintoview.Finallyweaddedalightsourceabovetheburr 21

PAGE 22

tosimulatethesmalllightthatissometimeslocatedatthebottomofahandpiece.Thislightcreatesrealisticshadowsandhighlightsonthetoothsurfacethatprovideusefulorientationinformationtotheuser.TheseorientationfeaturesareoverviewedinFigure 3-9 .Whentheviewofthetoothsurfaceisobscuredbythehandpiece,orwhenthetoothisotherwisenotindirectviewofthepractitioner,amirrorisused.CEDShasamirrorthatcanbetoggledonandoff. 3.5Conguration Figure3-7. GLConsoleInterface WeusetheGLConsolelibrarytoprovidesimpledynamiccongurationofparameters.Theseparameterscanbesavedandloaded,tohelpwithcustomizingvariablesusingdomainexperts.Thisconsoleisimplementedasatransparentoverlayoverthescene.Anyvariablesregisteredwiththecongurationmodulecanbeeditedinreal-timetoallowquicktestingofnewsettings. 3.6HapticsOuroriginalnaivehapticapproachwastousesurfacenormalstodirectlyspecifydeviceforces.Whilethisapproachcouldconceivablyworkforsoft,deformableobjects, 22

PAGE 23

Figure3-8. CEDShapticloopSequenceDiagram itdoesnottranslatewelltohardobjectssinceitdoesnotimposeconstraintsonthepositionstowhichthehandpiececanmove.Itoperatesundertheassumptionthatexertingforcesawayfromthesurfacewillprohibitthevirtualdevicefrompassingthroughillegalpositions.Whilethisapproachmightworkwithaveryhighresolutiondatastructureandanextremelypowerfuldevice,theNovintFalconRisnotpowerfulorpreciseenoughtoassumethatsuchanaiveapproachcouldcreatearealisticfeelingwhendraggingovercomplicatedtopography.Williametal.similarlyfoundthatsuchtechniquesperformerraticallyintightcornersorinsidepits[ 12 ].Toachievearealistic,hardfeelweusetheanchorapproach[ 12 ].Theanchoristhepointinthevirtualscenewheretheviewerwouldexpectthevirtualburrtoendupaccordingtorealphysics,ratherthantheactualdeviceposition.Thehandpiecewillalwaysbepulledtowardstheanchorwithaspringforce.Whenthevirtualburrisjudgedtobewithinthevirtualtooth,theanchormustbedraggedoverthetoothsurface,ratherthanintoit.Whenthehapticloopexecutesanewanchorpointischosenfromamong27anchorcandidates(the26adjacentpointsplusthecurrentpoint).Ouralgorithmchoosesthecandidateclosesttothevirtualdevicepositionthatdoesnotintersectwiththevirtualtooth.Intersectionofthevirtualburrwiththevirtualtoothisevaluatedusingaburrmaskdatastructure.WhileWilliametal.identiedtheanchorapproachwhenuseddirectlyonthevoxelmodelasblocky,ourapproachfeelsverysmoothonthesurface,andonlyexhibitssomeblockytendencieswherethemodelbecomesnearlyvertical. 23

PAGE 24

Theforcespullingthedevicetowardstheanchorwerecalibratedtobeashardaspossiblewithoutcausinganythrashingeffects.Theoutwardforce(inthezdirection)mustbescaledtopreventlateralthrashing,becausethehapticarmsoftheNovintFalconRarenotmechanicallycapableofworkingashardhorizontallyasvertically.Forconvenienceweintroduceacenteringstatetoautomaticallybringthedevicetothecenteroftheworkarea.Whileincenteringstatetheanchorpointwillmoveincrementallyclosertothecenterpoint.Centeringstopswhenthehandpieceispulledsufcientlyfarfromthecenteringanchor.Africtioneffectisvitalforarealistic,hardhapticresponse.Withoutfrictiontheuserpushingwithsufcientforcecantriggerathrashingeffectcreatedbyanunderdampedspringforce.Thefrictioneffectcreatesanover-dampedspringforcethatkeepsthedevicestableevenwhenpushedveryhard.Thisforceiscreatedbytrackingthemostrecentlocationduringthehapticloop.Thefrictionalforceisthedifferencebetweenthecurrentdevicepositionandthepreviousposition,scaledbyafrictionalfactorandclampedsothatthefrictioncannotovercometheforceproducedbytheuser'shandanddevolveintouncontrollablethrashing.Suddendiscontinuitiesinforceoccurbasedoninsufcienthapticrefreshspeedordatastructureresolution.Thesecanbesmoothedoutusingaforcelter.Thelterordamperclampsthemagnitudeofchangeinforceexertedbetweenconsecutivecallstothehapticloop[ 11 ].Wedonotallowthehandpiecetoexittheworkarea.Thishelpskeeptheuserfromlosingtrackofthevirtualhandpieceonthescreen.Ifthevirtualhandpieceismovedsufcientlyfaroutofthisareacenteringmodeisenabled.Whenthehandpieceisonbutnotincontactwiththetooth,theusershouldbeabletofeelthespinningburr.Thisisaccomplishedbyaddingtinyrandomizedforces,givingtheuseratactilesenseofthehandpiecebeingswitchedon. 24

PAGE 25

Conventionallyremovalusingaspinningburrismeasuredintermsofthefeedspeed,thespeedatwhichthespinningburrismovedintothematerialtoberemoved.Iftheburrismovedintothematerialsteadilyatorbelowtheacceptablerate,theusershouldfeellittleresistance.Pushingharderagainstthetoothwiththemovingburrshouldallowmoreremoval,uptosomelimitdependingonthetorqueoftheburr.Theabilitytoremovematerialisalsolimitedbytheareaandtypeofmaterialincontactwiththeburr.Finally,thedrillingdirectionneedstobeconsidered.Forinstanceatypicaldentaldrillcanremovemorematerialmoreeasilywhenmovedlaterallyratherthanvertically.Thefollowingarebasicdrillingrequirementsthatwegleanedfromdentalstudentsandprofessionals: Whenholdingamovinghighspeedhandpiece,iftheburraccidentallybrushesatooth,nonoticeabledamageshouldoccur.Openingupapreparationrequiressustainedpressure. Consistentremovalwithahighspeedburrshouldbepossiblewitharelativelylowforce(ontheorderofoneNewton). Toothmaterialshouldbenaturallyremovedbybrushingtheburrbackandforth.Holdingtheburrlightlyonthetoothsurfaceandbrushingitbackandforthshouldcreateatrenchthattheburrcannotslideoutofwithin10strokes.Wemeasuretheabilityofaspinningburrtocutintothetoothsurfacebyexamininganchorcandidatesthatcollidewiththetoothandweightingthemproportionallytotheirhardness.Thetotalvaluethatwouldbedecrementedfromalltoothheightmappointsintersectingwiththepotentialburriscalculated,scaledbythedistanceofthevirtualdevicetipfromtheanchorpoint(whichdescribestheforceexertedbytheuser),andcomparedtoaremovalthresholdcorrespondingtothedirectionthevirtualburrismoving.Anchorcandidatesthatwouldtranslatetheanchorsidewayshavealowerremovalthresholdthanverticalanchorcandidates,becauseaspinningburrtipismorecapableofremovingmateriallaterallythanvertically.Noticethattheremovalthresholdforanchorpointstranslatingthevirtualburrupwardisirrelevant. 25

PAGE 26

Whilewecansmoothlypushtheheightmapsurfacedowntocreateasatisfying,gradualremovaleffect,lateralremovalonourheightmapcanbeproblematic.Toproperlymanagethelateralremovalratewemustintroducearemovaltimer.Evenifthedrillingconstraintsseemtobemet,thetimermustrebeforeremovalmayoccuragaintoachieveasmootherlateralremoval. ADrillinghandpiece BRemovedcariesFigure3-9. Demonstrationofdrilling Collisiondetectionisachievedbycomparingacollectionofheightoffsetsrepresentingtheshapeoftheburrtothetoothsurfacearoundthevirtualdevicetip.Thispre-computedburrmaskavoidstheneedtoconstantlyrecomputethegeometryoftheburrshape.Theimplementationofthismaskisacombinationoftwoheightfunctionswhichisessentiallyidenticaltotheimplementationoftheunderlyingtoothdatamodel.Theburrisconsideredtobecollidingwithtoothiftheburrtopisabovethetoothbottom,oriftheburrbottomisbelowthetoothtop. 3.7SoundArealistichandpiecesoundplayswhenthehandpieceison.Improvingthesoundmodulewillbeimportantincreatingamoreeffectivemulti-modaldentaltrainingexperience. 26

PAGE 27

Figure3-10. CEDShapticloopActivityDiagram 3.8ErgonomicsItwasimportantforustocreateaconvincingergonomicsimulationusinglow-costsupplies.OurultimategoalistominimizetheamountoftimefromwhenanexperienceddentalpractitionerrstsitsdownwithCEDS,andwhenthepractitionerisabletoworkwithitnaturally.Adentalprofessionalshouldbeaninstantexpertatoperatingaproperdentalsimulationprogram.WeoriginallyattachedourAugmentedReality(AR)markertoamask(similartowhatapracticingdentistwouldwearduringworkonapatient),butthiswastoocumbersomeforoursimulationandintroducedsanitationissues(weneededtoprepare 27

PAGE 28

anothermaskforeverynewuser).Ourcurrentsolutionistoafxthemarkertoaheadband.Themarkerisonlydetectedifitiscompletelyunobscured,whichcanbetroublesomeforuserswithlongerhair. Figure3-11. AnARmarkerattachedtoaheadband TheballgripandpistolgripthatcomewiththeNovintFalconRdonotprovidearealisticinterfacetoadentalsimulation.Wewereabletoacquireabrokendentalhandpiecethatweattachedtoaballgrip.Whilethisisanexcellentwaytorecycleinoperabledrills,ifnohandpiececanbeobtainedapenortoothbrushcanservethispurpose. Figure3-12. AdentalhandpieceattachedtoaNovintFalconRgrip Werstchosetoreorientthedevicesothatthedevicearmspointupward.Consideringthattheprevailingforcefromtheuserwillbedownwardfordrilling,wewantedthedevicetobeorientedtoprovidethemostpowerfulandstableresponse.Theworkingareaishapticallyrenderedapproximatelyinthemiddleofthedevicecoordinatesystemtomakesuretheresponseismosteffective. 28

PAGE 29

Figure3-13. TheNovintFalconRwithattachedhandpieceorientedvertically Weliftthedeviceslightlyabovetheoorsothatwhenthedevicearmsareoutstretchedtothecenteringposition,thehandsofaseateduserlandcomfortablyonthehandpiece.Thedeviceisinaboxsothataatsurfacecanbeusedasafulcrumpointforthehand.Havingaspottorestthehandwhileoperatingthehandpieceisveryimportantfordentalpreparation.Itavoidsrequiringtheusertosuspendtheirarminplaceforlongperiodsoftime.Duringarealpreparation,adentistrestsanoutstretchedngeronanothertooth.Toswitchbetweenright-handedandleft-handedusers,thedevicecanbeturnedandtheviewcanbeipped.Figure 3-14 showsacompleteexperimentalsetup.TheARmarker,fulcrumsurface,handpiece,anddisplaywindowarehighlightedinred.While3DscreensorwearableVirtualRealitydisplaysareidealforimmersivesimulation,thecurrentgoalofCEDSisaneffectivetrainingsimulationusingubiquitousoff-the-countertechnology.Weplaceaconventionalscreeninbetweentheeyesoftheuserandthehandpieceinanattempttorenderthevirtualsceneinapproximatelythesameplacewheretherealtoothwouldbeexpected.Thissetupcanbeaccomplishedwithmostatscreens,includinglaptops(ifthescreenisippedupside-down). 29

PAGE 30

Figure3-14. StudentsattheUFCollegeofDentistrytryingCEDS 30

PAGE 31

CHAPTER4TESTINGANDEVALUATIONCEDSwastestedonUFdentalstudents.Theyweregivenabriefsetofinstructionsandwereinstructedtotryallthefeaturesofthesystem.Afterusingthesystemthestudentswereaskedtorespondtoabriefsurvey.TheinstructionsandasummaryofthesurveyresultsareincludedintheAppendix.Ingeneralstudentsandteachersreceivethesystementhusiastically,butagreethatitisnotquitereadytoreplaceconventionaltrainingmethods.ThissentimentissharedbyUFDentalProfessorKarl-JohanSoderholm. Thekeyweaknessisatrue3Dperception,whichmakesitdifculttoinitiallyorienttheburrinspace.Thetactilefeelingisnowthere,buteventhatneedstobereneduntilthesimulatorhasreacheditsfullestpotential.Whenaskedhowmuchastudentwouldpayforthesoftware(separatefromanyhardwarecosts)theaverageresponsewas$36.36(takingnotsure/notmuchas$0.00).ThoughonlyaminorityofrespondentsfoundthatCEDScouldreplaceconventionaltrainingmethods,amajorityimaginedthesystem,asitisnow,usefulforbeginningstudents.Testersratedorientationfeaturesliketheredlineextendingfromtheburr,thelightontheendofthehandpiece,andtheARmarkernavigationhighly.Theyagreedthatndingthevirtualtoothwiththehandpieceiseasy,evenwithoutareal3Dscreen.Additionalfeaturerequestsincludedamodeforalowspeedhandpiece,specialfeaturestoviewexactlyhowdeeptheuserisinthetooth(howclosetothepulpchamber),soundthatdynamicallychangeswithdrillspeedandremoval,alternativeburroptions,real3D,andafootpedalforcontrolofspeed. 31

PAGE 32

CHAPTER5FUTUREWORKAs3Ddisplaydevicesbecomemoreubiquitous,itmightnolongertooexpensivetouseasmallglasses-free3Dscreen(perhapsacellularphoneorportablegamingdevice).Wemayalsotestouttheeffectivenessofsimplered/blue3D(whichdoesnotrequireaspecialscreen),andinparticularthecombinationoftheARmarkerinputwith3Dglasses.Ifthescreencouldbeattacheddirectlytothehapticdevice,thenthesimulationsetupwouldbecometrulyportable.Weimaginestudentsbeingabletocheckoutahapticdevicefromtheirtraininglabandpracticeontheircomputerathome. Figure5-1. Anaglyph3Dred/blueglasseswiththeCEDS'UF'ARMarkerAttached Becausethedatastructureisrendereddirectlyasaheightmapitisefcientandcanbefurtheroptimized(perhapsthroughvertexarraysorGPUprogramming).Theentireapplicationcouldthenbeexecutedonaportabledevice.ThenextversionofCEDSmightbemadetorunona3Dgamingdeviceorsmartphone.Itshouldbeeasytoaddundocapabilitiessothatuserscanstepbacktoapreviousversionoftheirpreparationandtryagain.Wecankeepasmanysnapshotsofthedatamodelasmemorywouldallowandswapbetweenthemwhendesired.Undosemanticswouldgohand-in-handwithaninfractionsystemthatwoulddetectwhentheusermakesamajormistakeandgivetheusertheoptionofgoingbackandtryingagain.Aweaknessofthecurrentdatamodelisthatthereisnoconsiderationofthetooth'sstructuralintegrity.Inrealitywhendentistsprepareatooththeymustbeverycarefulnottoleavethedrilledareabrittle.Furthermore,anysmallprotrusionswouldlikelysnapoff 32

PAGE 33

becausetheyaretoobrittle.Inthecurrentimplementationitwouldbepossibletocreatetall,thincolumnsthatwouldbeimpossibletoconstructbydrillingonarealtooth.Inthefuturewecanaddasmoothingstepthatwouldcheckthelocaldrillingareaforsuchimpossiblestructuresandremovethem.Toothdecayandtoothlayergenerationcanbeslightlyrandomized.Insteadofsimplyalwaysllingupthelowestpartsofthetoothsurfacewithdecay,decaywouldbeimplementedasasemi-randomwalkwithapossibletendencytofollowthevarioustoothsurfacegrooves.Thedepthsandhardnessofthedecayandtoothlayerscouldalsobeslightlyrandomizedtomakesurethatstudentsarepreparedfortheunexpected.Soundpitchshouldbeadjustedtoreacttotheburrspeed.Soundshouldalsoreacttomaterialcontact.Wewouldliketondafootpedaltocontrolburrspeed.Thishasprovendifcultbecausecurrentlyavailablecommercialfootpedalsareeitherbinaryswitches,ightpedals,orattachedtoamassivesteeringwheelassembly.Wewouldliketobeabletoprovidemoreinterestinggraphicalfeedbacktotheusertoenhancethelearningexperiencefornovices.Thiscouldincludegeneratingasimulationofanx-rayimageofthetoothsothatthestudentcouldpracticeunderstandinghowtodrillatoothusingthesameinformationadentistmightreceiveinarealworkenvironment.Beginningstudentsmightbenetfromseeingareal-timecrosssectionoftheirdrillingworktogetabettersenseofhowfardowntheycandrillbeforegoingtoodeepandpenetratingthepulpchamber.Wewouldliketodrawafullmouthofteethatoncetohelporienttheuserandsimplythetoothchoosingprocess.Wewouldalsoliketorendertherestoftheoralcavityandexternalmouthtofullysurpasstherealismofthetypodont.Additionally,wewanttoallowtheorientationofthemouthtochangerelativetothehandpiece.Finally,inthefuturewewillallowdentalexpertstocalibratethepreparationgrading.Similartohowwecantesthowclosetheuseristoremovingalldecay,wecouldtestto 33

PAGE 34

seehowclosetheuseristomatchingtheidealpreparation.Allowingausertoeasilycomparetheirworktotheidealpreparationandretryasmuchaspossiblywouldcreatetheultimatetoolfordentaltraining. 34

PAGE 35

APPENDIXAUSERTESTINGINSTRUCTIONS A.1IntroductionCEDSisasimpleandcost-effectivedentaldrillingsimulationthatusescommonoff-the-shelftechnologyinanattempttoreplacetraditionaldentalpreparationtrainingmethods.WhenusingCEDStrytositandgraspthehandpiecenaturally.Movingtherealhandpiecewillmovethehandpieceonthescreen.Youwillactuallyfeelarealforcewhenyoutouchthevirtualtoothwiththevirtualhandpiece.Thehardnessofthevirtualtoothmaterialandthespeedofremovalcanbeadjustedatanytime. A.2Controls Grade(g):Pressthe'g'keytodisplaythepercentageofdecayremaining. Reload(r):Pressthe'r'keytoreloadthetooth(undoalldrilling). Drilling(d):Pressthe'd'keyorthespacebartotoggledrillingon/off(comparedrillingmodetoahigh-speed330carbidebur). Hand(h):Pressthe'h'keytoswitchbetweenaright-handedviewandaleft-handedview. Viewing(withAugmentedReality):Wearaheadbandwiththeblocky'UF'logo.ThisisanAugmentedReality(AR)marker.Ifthecamerathatisonthetopofthecomputerscreenhasagoodviewofthismarker,thenthevirtualscenewillrotaterelativetothepositionofyourhead.Usethistooltohelpyounaturallynavigatethescene. Viewing(withthemouse):Clickanddragonthecomputertrackpadtorotatethevirtualscene. Zooming:Puttwongersonthetrackpadanddragupordowntozoominoroutofthevirtualscene. SwitchingTeeth:Thevirtualteethareallloadedfrom3Dscansofrealteeth.21-28and31-38areavailable(FDInumbering).Thedefaultis36. 35

PAGE 36

Decay:Theblackdecayisautomaticallygeneratedbasedonthe3Dmodelofthetooth.Thedecayisplacedbyndingthelowestpointonthetoothclosesttothetoothcenter(whenlookingfromabove).Thedecaywillextenddownwardthroughtheenamel.Inthefuturethedecaywouldberandomizedtoenhancetrainingeffectiveness. Dentin:Dentiniscoloredyellowandlocatedataconstantdepthbelowtheenamelsurface.Inthefuturethedentinwouldberandomizedtoenhancetrainingeffectiveness. 36

PAGE 37

APPENDIXBSURVEYRESULTSTheseresultsarefrom11UFDentalStudents(4Juniors,7Seniors). Aforcefeedbacksystemcostsapproximately$150.Howmuchinadditiontothecostoftheforcefeedbacksystem,wouldyoubewillingtospendtoownacopyofCEDS? $0/NotSure/NotMuch:8 $50:1 $150:1 $200:1 WheredoyouseeCEDSttingintothedentalcurriculum? Firstyearorpre-dental:8 Idon'tseeitttingintothedentalcurriculum/notsure:2 Practicingbeforecuttingaplastictooth/fororientationonly,notsimulation:3 Whatextrafeatureswouldbeuseful?(optional) Lowspeedhandpiece:3 Cross-section/X-rayview:1 Drillsoundpitchmodulation:2 Moreburrs:2 Real3D:1 FootPedal:1 DoesCEDShavemajorshortcomings?Howcantheybeimproved?(optional) Difculttoorient:1 37

PAGE 38

TableB-1. Quantitativesurveyresponses StronglyDisagreeDisagreeDon'tKnowAgreeStronglyAgree Thevirtualtoothlookslikearealtooth.0.0%(0)0.0%(0)0.0%(0)54.5%(6)45.5%(5)Thevirtualtoothfeelslikearealtooth.27.3%(3)36.4%(4)0.0%(0)36.4%(4)0.0%(0)Materialremovalfromthetoothfeelsplausible.0.0%(0)54.5%(6)9.1%(1)36.4%(4)0.0%(0)CEDScouldbeusedinlieuofthedentalpracticepad.9.1%(1)36.4%(4)27.3%(3)27.3%(3)0.0%(0)CEDScouldbeusedinlieuofplasticpracticeteeth.27.3%(3)63.6%(7)0.0%(0)9.1%(1)0.0%(0)CEDScouldbeusedinlieuofrealteeth.81.8%(9)18.2%(2)0.0%(0)0.0%(0)0.0%(0)Findingthevirtualtoothwiththehandpieceiseasy.0.0%(0)27.3%(3)9.1%(1)54.5%(6)9.1%(1)Theverticalredlinehelpspositiontheburronthevirtualtooth.0.0%(0)0.0%(0)0.0%(0)45.5%(5)54.5%(6)Thelightcomingfromthehandpiecehelpsorientatetheburraroundthevirtualtooth.0.0%(0)0.0%(0)0.0%(0)72.7%(8)27.3%(3)The'UF'augmentedrealitymarkerhelpsorienttheview.9.1%(1)9.1%(1)9.1%(1)45.5%(5)27.3%(3)CEDSshouldbepartofeverydentalstudent'straining.18.2%(2)0.0%(0)72.7%(8)9.1%(1)0.0%(0) 38

PAGE 39

REFERENCES [1] W.Mix,Learn-a-prepii. http://whipmix.com/product/learn-a-prep-ii/ [2] L.Boushell,R.Walter,andC.Phillips,Learn-a-prepiiasapredictorofpsychomotorperformanceinarestorativedentistrycourse,JournalofDentalEducation,vol.75,no.10,pp.1362,2011. [3] K.H.,Artoolkit. http://www.hitl.washington.edu/artoolkit/ [4] C.Dentoform,Simulationteeth. http://www.columbiadentoform.com/main.asp?CAT=5 [5] G.Thomas,L.Johnson,S.Dow,andC.Stanford,Thedesignandtestingofaforcefeedbackdentalsimulator,ComputerMethodsandProgramsinBiomedicine,vol.64,no.1,pp.53,2001. [6] L.KimandS.Park,Hapticinteractionandvolumemodelingtechniquesforrealisticdentalsimulation,TheVisualComputer:InternationalJournalofComputerGraphics,vol.22,pp.90,2006. [7] H.Noborio,D.Sasaki,Y.Kawamoto,T.Tatsumi,andT.Sohmura,Mixedrealitysoftwarefordentalsimulationsystem,inHapticAudiovisualEnvironmentsandGames,InternationalWorkshopon,pp.19,2008. [8] P.Rhienmora,P.Haddawy,P.Khanal,S.Suebnukarn,andM.Dailey,Avirtualrealitysimulatorforteachingandevaluatingdentalprocedures,MethodsofInformationinMedicine,vol.49,pp.396,2010. [9] G.Liu,Y.Zhang,andW.Townsend,Forcemodelingfortoothpreparationinadentaltrainingsystem,VirtualReality,vol.12,pp.125,2008. [10] J.Wu,G.Yu,D.Wang,Y.Zhang,andC.Wang,Voxel-basedinteractivehapticsimulationofdentaldrilling,inProceedingsoftheASME2009InternationalDesignEngineeringTechnicalConferencesComputersandInformationinEngineeringConference,pp.1,2009. [11] H.Yau,L.Tsou,andM.Tsai,Octree-basedvirtualdentaltrainingsystemwithahapticdevice,Computer-AidedDesignApplications,vol.3,pp.415,2006. [12] J.William,O.Telles,andL.W.S.,Interactive3dhapticcarvingusingcombinedvoxelsandmesh,inHapticAudioVisualEnvironmentsandGames,InternationalWorkshopon,pp.108,2008. [13] Novint,Virtualrealitydentaltrainingsystem(vrdts). http://www.novint.com/index.php/medicaldental [14] Sensable,Sensabledental. http://www.sensabledental.com/dental-home.htm [15] VOXEL-MAN,Voxel-mandental. http://www.voxel-man.de/simulator/dental/ 39

PAGE 40

[16] Moog,Haptictechnologyinthemoogsimodontdentaltrainer. http://www.moog.com/markets/medical-dental-simulation/haptic-technology-in-the-moog-simodont-dental-trainer/ [17] I.Navigation,Dentsim. http://www.denx.com/DentSim/overview.html [18] M.Nishida,T.Sohmura,andJ.Takahashi,Trainingintoothpreparationutilizingasupportsystem,JournalofOralRehabilitation,vol.31,pp.149,2004. 40

PAGE 41

BIOGRAPHICALSKETCH WilliamChristopherHuntreceivedhisB.S.inDigitalArtsandSciencesandhisB.A.inEastAsianLanguagesandLiteraturesfromtheUniversityofFloridain2010,wherehegraduatiedwithhonors.Hespentthe2008-2009academicyearasanundergraduateinTokyoatAoyamaGakuinUniversitythroughaUFexchangeprogram.WilliambeganresearchwithJorgPetersintheUFSurfLabin2010asagraduatestudent,wherehefocusedontheCEDSdentalproject.Duringhisstudies,WilliamworkedasasoftwaredeveloperattheGainesvilletestpreparationcompanyGleimPublicationsstartingin2008.In2011,heacceptedapositionasaSystemsEngineerwithMitsubishiHeavyIndustriesinYokohama,Japan. 41