2024
Anatomically and Metabolically Informed Deep Learning Low-Count PET Image Denoising
Xia M, Xie H, Liu Q, Guo L, Ouyang J, Bayerlein R, Spencer B, Badawi R, Li Q, Fakhri G, Liu C. Anatomically and Metabolically Informed Deep Learning Low-Count PET Image Denoising. 2024, 00: 1-2. DOI: 10.1109/nss/mic/rtsd57108.2024.10657099.Peer-Reviewed Original ResearchDeep learningOver-smoothed imagesDL training processesHigh-count imagesImage denoisingDenoised imageLow-count dataSemantic informationSemantic classesSegmentation guidanceTraining processPET/CT systemHistogram distributionImage qualitySegmentation toolPositron emission tomographyImagesDenoisingDatasetHistogramPriorsRadiation exposurePopulation-based deep image prior for dynamic PET denoising: A data-driven approach to improve parametric quantification
Liu Q, Tsai Y, Gallezot J, Guo X, Chen M, Pucar D, Young C, Panin V, Casey M, Miao T, Xie H, Chen X, Zhou B, Carson R, Liu C. Population-based deep image prior for dynamic PET denoising: A data-driven approach to improve parametric quantification. Medical Image Analysis 2024, 95: 103180. PMID: 38657423, DOI: 10.1016/j.media.2024.103180.Peer-Reviewed Original ResearchDeep Image PriorImage priorsSupervised modelsNoise reductionIntrinsic image featuresDeep learning techniquesU-Net architectureNovel denoising techniqueQuality of parametric imagesDenoising modelDenoising techniquesStatic datasetsBaseline techniquesEffective noise reductionData-driven approachLearning techniquesDynamic datasetsOptimization processPrior informationStatic imagesHigh noise levelsImage featuresDatasetPrior imagePET datasets
2022
Virtual high‐count PET image generation using a deep learning method
Liu J, Ren S, Wang R, Mirian N, Tsai Y, Kulon M, Pucar D, Chen M, Liu C. Virtual high‐count PET image generation using a deep learning method. Medical Physics 2022, 49: 5830-5840. PMID: 35880541, PMCID: PMC9474624, DOI: 10.1002/mp.15867.Peer-Reviewed Original ResearchConceptsStructural similarity indexImage quality evaluationDeep learning-based methodsDeep learning methodsImage qualityLearning-based methodsPET datasetsStatic datasetsDL methodsNet networkImage generationPET imagesNetwork inputsImage counterpartsLearning methodsNetwork outputTraining datasetPeak signalPositron emission tomography (PET) imagesQuality evaluationDatasetCross-validation resultsMean square errorHigh-count imagesImagesPET respiratory motion correction: quo vadis?
Lamare F, Bousse A, Thielemans K, Liu C, Merlin T, Fayad H, Visvikis D. PET respiratory motion correction: quo vadis? Physics In Medicine And Biology 2022, 67: 03tr02. PMID: 34915465, DOI: 10.1088/1361-6560/ac43fc.Peer-Reviewed Original ResearchConceptsRespiratory motion correctionPET respiratory motion correctionMotion correctionGeneric motion modelImage reconstruction processRespiratory motion informationMotion estimationMotion informationTerms of synchronizationImage spaceSynchronization stepsReconstruction processMotion modelOverall approachPET/magnetic resonance imagingDevice systemRespiratory motionImaging devicesMRI deviceDevicesSynchronizationNumber of stepsComprehensive coverageDatasetGreat interest
2021
Investigation of Direct and Indirect Approaches of Deep-Learning-Based Attenuation Correction for General Purpose and Dedicated Cardiac SPECT Scanners
Chen X, Zhou B, Xie H, Shi L, Liu H, Liu C. Investigation of Direct and Indirect Approaches of Deep-Learning-Based Attenuation Correction for General Purpose and Dedicated Cardiac SPECT Scanners. 2021, 00: 1-2. DOI: 10.1109/nss/mic44867.2021.9875517.Peer-Reviewed Original ResearchNovel neural networkConventional U-NetMulti-channel inputDeep learningU-NetAttenuation mapNeural networkMap generationCardiac SPECTGeneral purposeSuperior performanceImagesDatasetIterative reconstructionAttenuation-corrected imagesCT transmission scanningAveraged errorNovel methodParallel-hole SPECTAttenuation correctionSPECT scannerMapsEmission imagesDirect approachScannerLimited View Tomographic Reconstruction Using a Cascaded Residual Dense Spatial-Channel Attention Network With Projection Data Fidelity Layer
Zhou B, Zhou S, Duncan JS, Liu C. Limited View Tomographic Reconstruction Using a Cascaded Residual Dense Spatial-Channel Attention Network With Projection Data Fidelity Layer. IEEE Transactions On Medical Imaging 2021, 40: 1792-1804. PMID: 33729929, PMCID: PMC8325575, DOI: 10.1109/tmi.2021.3066318.Peer-Reviewed Original ResearchConceptsAttention networkView reconstructionGrand challenge datasetLimited angle reconstructionHigh-quality reconstructionNeural network methodSparse-view reconstructionExperimental resultsLimited angle acquisitionArchitecture issuesSparse viewsChallenge datasetLimited view dataView dataNeural architectureQuality reconstructionNetwork methodTomographic reconstructionReconstructed imagesProjection viewsPrevious methodsAngle reconstructionDatasetNetworkAngle acquisition