2022
Incremental Learning Meets Transfer Learning: Application to Multi-site Prostate MRI Segmentation
You C, Xiang J, Su K, Zhang X, Dong S, Onofrey J, Staib L, Duncan J. Incremental Learning Meets Transfer Learning: Application to Multi-site Prostate MRI Segmentation. Lecture Notes In Computer Science 2022, 13573: 3-16. PMID: 37415747, PMCID: PMC10323962, DOI: 10.1007/978-3-031-18523-6_1.Peer-Reviewed Original ResearchIncremental learningMedical image segmentation tasksMulti-site datasetImage segmentation tasksMedical image segmentationProstate MRI SegmentationComputation resourcesMedical datasetsSegmentation taskImage segmentationSegmentation frameworkEmbedding featuresBenchmark datasetsMRI segmentationTraining dataTarget domainLearning approachPractical deploymentDomain-specific expertiseCompetitive performanceDatasetTraining schemePrior workSegmentationSingle modelSimultaneous Segmentation and Motion Estimation of Left Ventricular Myocardium in 3D Echocardiography Using Multi-task Learning
Ta K, Ahn SS, Stendahl JC, Langdon J, Sinusas AJ, Duncan JS. Simultaneous Segmentation and Motion Estimation of Left Ventricular Myocardium in 3D Echocardiography Using Multi-task Learning. Lecture Notes In Computer Science 2022, 13131: 123-131. PMID: 35759335, PMCID: PMC9221412, DOI: 10.1007/978-3-030-93722-5_14.Peer-Reviewed Original ResearchMotion estimationMulti-task learning networkMedical image analysis literatureMulti-task learningSingle-task learningMotion estimation techniqueImage analysis literatureComputer visionDecoding branchesFeature encoderLearning frameworkLearning networkLatent featuresAccurate segmentationSimultaneous segmentationEstimate motionImage pairsTask learningRealistic motion patternsVolumetric segmentationSegmentationMotion patternsTaskUnique taskLearning
2021
Anatomy-Constrained Contrastive Learning for Synthetic Segmentation Without Ground-Truth
Zhou B, Liu C, Duncan J. Anatomy-Constrained Contrastive Learning for Synthetic Segmentation Without Ground-Truth. Lecture Notes In Computer Science 2021, 12901: 47-56. DOI: 10.1007/978-3-030-87193-2_5.Peer-Reviewed Original ResearchSegmentation networkContrastive learningManual segmentationSuperior segmentation performanceObject of interestSynthetic SegmentationManual effortSegmentation performanceTraining dataUnsupervised adaptationImaging dataSource modalitySegmentationNetworkPrevious methodsLearningLarge amountSuccessful applicationPET imaging dataImagesObjectsCodeDataNew imaging modalityMulti-frame Attention Network for Left Ventricle Segmentation in 3D Echocardiography
Ahn SS, Ta K, Thorn S, Langdon J, Sinusas AJ, Duncan JS. Multi-frame Attention Network for Left Ventricle Segmentation in 3D Echocardiography. Lecture Notes In Computer Science 2021, 12901: 348-357. PMID: 34729554, PMCID: PMC8560213, DOI: 10.1007/978-3-030-87193-2_33.Peer-Reviewed Original ResearchPerformance of segmentationLeft ventricle segmentationVentricle segmentationMedical image segmentation modelsSpatiotemporal featuresAttention networkImage segmentation modelSequence of imagesAttention mechanismSegmentation modelTedious taskTarget imageSegmentationEchocardiography imagesExperimental resultsImagesNetworkShape-Regularized Unsupervised Left Ventricular Motion Network With Segmentation Capability In 3d+ Time Echocardiography
Ta K, Ahn SS, Stendahl JC, Sinusas AJ, Duncan JS. Shape-Regularized Unsupervised Left Ventricular Motion Network With Segmentation Capability In 3d+ Time Echocardiography. 2011 IEEE International Symposium On Biomedical Imaging: From Nano To Macro 2021, 00: 536-540. PMID: 34168721, PMCID: PMC8221369, DOI: 10.1109/isbi48211.2021.9433888.Peer-Reviewed Original ResearchConvolutional neural networkAccurate motion estimationCardiac motion patternsMotion estimation performanceDense displacement fieldB-mode echocardiography imagesSegmentation masksMedical imagesMotion estimationNeural networkSegmentation capabilityTarget imageUnsupervised estimationImportant taskSegmentationMotion patternsDisplacement fieldNetworkEchocardiography imagesEstimation performanceImagesLow signalAdditional challengesMotion networkNoise ratioAnatomy-guided multimodal registration by learning segmentation without ground truth: Application to intraprocedural CBCT/MR liver segmentation and registration
Zhou B, Augenfeld Z, Chapiro J, Zhou SK, Liu C, Duncan JS. Anatomy-guided multimodal registration by learning segmentation without ground truth: Application to intraprocedural CBCT/MR liver segmentation and registration. Medical Image Analysis 2021, 71: 102041. PMID: 33823397, PMCID: PMC8184611, DOI: 10.1016/j.media.2021.102041.Peer-Reviewed Original ResearchConceptsMultimodal registrationLiver segmentationLarge-scale manual annotationGround truthMultimodal image registrationMultimodal registration methodSegmentation networkDomain adaptationManual annotationSource modalityImage registrationRegistration frameworkSegmentationImage-guided interventionsRegistration methodMedical imagingDiagnostic medical imagingCorrect transformationLimited FOVStructure informationIntraprocedural CBCTImage qualitySegmenterExperimental resultsPatient data
2020
Sparse Data–Driven Learning for Effective and Efficient Biomedical Image Segmentation
Onofrey JA, Staib LH, Huang X, Zhang F, Papademetris X, Metaxas D, Rueckert D, Duncan JS. Sparse Data–Driven Learning for Effective and Efficient Biomedical Image Segmentation. Annual Review Of Biomedical Engineering 2020, 22: 1-27. PMID: 32169002, PMCID: PMC9351438, DOI: 10.1146/annurev-bioeng-060418-052147.Peer-Reviewed Original ResearchUnsupervised motion tracking of left ventricle in echocardiography
Ahn SS, Ta K, Lu A, Stendahl JC, Sinusas AJ, Duncan JS. Unsupervised motion tracking of left ventricle in echocardiography. Proceedings Of SPIE--the International Society For Optical Engineering 2020, 11319: 113190z-113190z-7. PMID: 32994659, PMCID: PMC7521020, DOI: 10.1117/12.2549572.Peer-Reviewed Original ResearchMotion trackingGround truth displacement fieldsConvolutional neural networkAccurate motion trackingDense displacement fieldB-mode echocardiography imagesU-NetNeural networkTracking frameworkNon-rigid registration algorithmTarget imageRegistration algorithmTarget frameSource frameAlgorithmEchocardiography imagesFavorable performanceDatasetImagesTrackingDisplacement estimationLarge amountEchocardiographic imagesSegmentationNetwork
2019
Domain-Agnostic Learning with Anatomy-Consistent Embedding for Cross-Modality Liver Segmentation
Yang J, Dvornek NC, Zhang F, Zhuang J, Chapiro J, Lin M, Duncan JS. Domain-Agnostic Learning with Anatomy-Consistent Embedding for Cross-Modality Liver Segmentation. ICCV Workshops 2019, 00: 323-331. PMID: 34676308, PMCID: PMC8528125, DOI: 10.1109/iccvw.2019.00043.Peer-Reviewed Original ResearchDomain adaptationDisentangled representationsLiver segmentationTarget domainSource domainDeep learning modelsGenerative adversarial networkHuman interpretabilityLearning frameworkAdversarial networkDownstream tasksArt methodsSegmentation consistencyLearning modelAgnostic learningMeaningful representationCycleGANNew tasksAblation analysisDA taskDifferent modalitiesTaskSegmentationEmbeddingLearning
2011
Segmentation of 3D radio frequency echocardiography using a spatio-temporal predictor
Pearlman PC, Tagare HD, Lin BA, Sinusas AJ, Duncan JS. Segmentation of 3D radio frequency echocardiography using a spatio-temporal predictor. Medical Image Analysis 2011, 16: 351-360. PMID: 22078842, PMCID: PMC3267850, DOI: 10.1016/j.media.2011.09.002.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsAnimalsComputer SimulationDogsEchocardiography, Three-DimensionalHeart VentriclesImage EnhancementImage Interpretation, Computer-AssistedImaging, Three-DimensionalModels, CardiovascularModels, StatisticalPattern Recognition, AutomatedReproducibility of ResultsSensitivity and SpecificitySubtraction TechniqueConceptsLeft ventricular endocardial boundarySpatio-temporal predictorsStandard level setRF dataSpatio-temporal coherenceNeighboring framesImage sequencesBoundary detectionMultiple framesImage inhomogeneitySegmentationEndocardial boundaryGeometric constraintsManual tracingRF ultrasoundAlgorithmLevel setsEchocardiographic imagesFrameConditional modelLinear predictorTrackingSpatial modelImagesRobustnessSegmentation of 3D RF Echocardiography Using a Multiframe Spatio-temporal Predictor
Pearlman PC, Tagare HD, Lin BA, Sinusas AJ, Duncan JS. Segmentation of 3D RF Echocardiography Using a Multiframe Spatio-temporal Predictor. Lecture Notes In Computer Science 2011, 22: 37-48. PMID: 21761644, DOI: 10.1007/978-3-642-22092-0_4.Peer-Reviewed Original ResearchConceptsLeft ventricular endocardial boundarySpatio-temporal predictorsStandard level setSpatio-temporal coherenceNeighboring framesImage sequencesBoundary detectionRF dataMultiple framesImage inhomogeneitySegmentationEndocardial boundaryGeometric constraintsManual tracingRF ultrasoundLevel setsConditional modelEchocardiographic imagesFrameLinear predictorAlgorithmTrackingSpatial modelImagesRobustness
2010
A coupled deformable model for tracking myocardial borders from real-time echocardiography using an incompressibility constraint
Zhu Y, Papademetris X, Sinusas AJ, Duncan JS. A coupled deformable model for tracking myocardial borders from real-time echocardiography using an incompressibility constraint. Medical Image Analysis 2010, 14: 429-448. PMID: 20350833, PMCID: PMC4318707, DOI: 10.1016/j.media.2010.02.005.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsAnimalsArtificial IntelligenceComputer SystemsDogsEchocardiography, Three-DimensionalElasticity Imaging TechniquesHumansImage EnhancementImage Interpretation, Computer-AssistedPattern Recognition, AutomatedReproducibility of ResultsSensitivity and SpecificitySubtraction TechniqueConceptsDeformable modelImage-derived informationLV endocardial boundariesImage acquisition techniquesFinal segmentationAutomatic algorithmGround truthManual segmentationVolumetric imagesSegmentationSynthetic dataEndocardial boundaryNumber of effortsMyocardial bordersEpicardial boundariesAcquisition techniquesInstantaneous acquisitionConstraintsImagesEchocardiographic imagesSetSpeckle statisticsAlgorithmReal-time echocardiography3D Radio Frequency Ultrasound Cardiac Segmentation Using a Linear Predictor
Pearlman PC, Tagare HD, Sinusas AJ, Duncan JS. 3D Radio Frequency Ultrasound Cardiac Segmentation Using a Linear Predictor. Lecture Notes In Computer Science 2010, 13: 502-509. PMID: 20879268, PMCID: PMC3889143, DOI: 10.1007/978-3-642-15705-9_61.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsAnimalsComputer SimulationDogsEchocardiography, Three-DimensionalImage EnhancementImage Interpretation, Computer-AssistedImaging, Three-DimensionalLinear ModelsModels, CardiovascularMyocardial InfarctionPattern Recognition, AutomatedRadio WavesReproducibility of ResultsSensitivity and SpecificityConceptsLeft ventricular endocardial boundaryStandard level setSpatio-temporal coherenceCardiac segmentationBoundary detectionImage inhomogeneityEndocardial boundarySegmentationGeometric constraintsManual tracingRadio frequency ultrasoundLinear predictorLevel setsRF dataEchocardiographic imagesB-mode dataTrackingImagesDataConstraintsSetDetectionIntegrated Segmentation and Nonrigid Registration for Application in Prostate Image-Guided Radiotherapy
Lu C, Chelikani S, Chen Z, Papademetris X, Staib LH, Duncan JS. Integrated Segmentation and Nonrigid Registration for Application in Prostate Image-Guided Radiotherapy. Lecture Notes In Computer Science 2010, 13: 53-60. PMID: 20879214, DOI: 10.1007/978-3-642-15705-9_7.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsHumansImaging, Three-DimensionalMaleProstatic NeoplasmsRadiographic Image EnhancementRadiographic Image Interpretation, Computer-AssistedRadiotherapy, Computer-AssistedReproducibility of ResultsSensitivity and SpecificitySubtraction TechniqueSystems IntegrationTomography, X-Ray ComputedConceptsManual segmentationAutomatic segmentationImportant treatment parametersNonrigid registrationImage-guided radiotherapy systemReal patient dataNon-rigid registrationIntegrated SegmentationRegistration partRadiotherapy linear acceleratorSegmentationTreatment imagesImage qualityCone-beam CTTreatment parametersImagesPromising resultsPatient dataKey anatomical structuresLinear acceleratorRegistrationPrevious workRadiotherapy system
2009
A Dynamical Shape Prior for LV Segmentation from RT3D Echocardiography
Zhu Y, Papademetris X, Sinusas AJ, Duncan JS. A Dynamical Shape Prior for LV Segmentation from RT3D Echocardiography. Lecture Notes In Computer Science 2009, 5761: 206-213. PMID: 20054422, PMCID: PMC2801876, DOI: 10.1007/978-3-642-04268-3_26.Peer-Reviewed Original ResearchSubject-specific dynamical modelCurrent frameMotion patternsRecursive Bayesian frameworkSegmentation taskPast framesAutomatic segmentationPrevious frameSegmentation processShape priorsLV segmentationManual segmentationSegmentationIntensity informationCardiac sequenceEchocardiographic sequencesStatic modelPrior knowledgeTemporal coherenceDynamical shape priorsCardiac motionCardiac modelsBayesian frameworkGeneric dynamical modelEchocardiographic imagesA dynamical shape prior for LV segmentation from RT3D echocardiography.
Zhu Y, Papademetris X, Sinusas AJ, Duncan JS. A dynamical shape prior for LV segmentation from RT3D echocardiography. 2009, 12: 206-13. PMID: 20425989, PMCID: PMC7814293.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsArtificial IntelligenceComputer SimulationComputer SystemsEchocardiography, Three-DimensionalHeart VentriclesHumansImage EnhancementImage Interpretation, Computer-AssistedImaging, Three-DimensionalModels, AnatomicPattern Recognition, AutomatedPhantoms, ImagingReproducibility of ResultsSensitivity and SpecificitySubtraction TechniqueConceptsSubject-specific dynamical modelCurrent frameMotion patternsRecursive Bayesian frameworkSegmentation taskPast framesAutomatic segmentationPrevious frameSegmentation processLV segmentationManual segmentationSegmentationIntensity informationCardiac sequenceEchocardiographic sequencesStatic modelPrior knowledgeTemporal coherenceCardiac motionCardiac modelsBayesian frameworkGeneric dynamical modelEchocardiographic imagesFrameInter-subject variability
2007
Segmentation of Myocardial Volumes from Real-Time 3D Echocardiography Using an Incompressibility Constraint
Zhu Y, Papademetris X, Sinusas A, Duncan JS. Segmentation of Myocardial Volumes from Real-Time 3D Echocardiography Using an Incompressibility Constraint. Lecture Notes In Computer Science 2007, 10: 44-51. PMID: 18051042, DOI: 10.1007/978-3-540-75757-3_6.Peer-Reviewed Original ResearchConceptsAutomatic segmentationImage-derived informationLV endocardial boundariesFinal representationManual segmentationSegmentationEndocardial boundaryEpicardial boundariesReal-time 3D echocardiographyTight couplingNew approachThree-dimensional shapeConstraintsVariety of effortsRepresentationInformationNew imaging modalitySetLEVEL SET BASED CLUSTERING FOR ANALYSIS OF FUNCTIONAL MRI DATA
Bathula DR, Papademetris X, Duncan JS. LEVEL SET BASED CLUSTERING FOR ANALYSIS OF FUNCTIONAL MRI DATA. 2011 IEEE International Symposium On Biomedical Imaging: From Nano To Macro 2007, 4: 416-419. PMID: 20216927, PMCID: PMC2834251, DOI: 10.1109/isbi.2007.356877.Peer-Reviewed Original ResearchContextual informationSpatio-temporal contextual informationLocal spatial contextBased ClusteringBetter segmentationReal fMRI experimentsSimilarity measureSynthetic imagesBrain imagesSimilar temporal behaviorTwo-dimensional curvesFunctional brain imagesSimulation resultsSpatial contextLevel setsMRI dataImagesTemporal behaviorSimilar approachInformationVoxelsSegmentationTime seriesFunctional MRI dataAdjacent voxelsLV Segmentation Through the Analysis of Radio Frequency Ultrasonic Images
Yan P, Jia CX, Sinusas A, Thiele K, O’Donnell M, Duncan JS. LV Segmentation Through the Analysis of Radio Frequency Ultrasonic Images. Lecture Notes In Computer Science 2007, 20: 233-244. PMID: 17633703, DOI: 10.1007/978-3-540-73273-0_20.Peer-Reviewed Original Research
2004
3D image segmentation of deformable objects with joint shape-intensity prior models using level sets
Yang J, Duncan JS. 3D image segmentation of deformable objects with joint shape-intensity prior models using level sets. Medical Image Analysis 2004, 8: 285-294. PMID: 15450223, PMCID: PMC2832842, DOI: 10.1016/j.media.2004.06.008.Peer-Reviewed Original ResearchConceptsImage segmentationImage gray levelsPoint distribution modelObject shapeGray levelsExplicit point correspondencesImage gray level valuesGray level valuesMedical imagesInput imageTraining imagesGray-level variationsMultidimensional dataTraining phaseDeformable objectsPoint correspondencesSegmentationMap shapePrior knowledgePrior informationLevel set functionPrior modelEstimation modelImagesObjects