2024
Point-supervised Brain Tumor Segmentation with Box-prompted Medical Segment Anything Model
Liu X, Woo J, Ma C, Ouyang J, Fakhri G. Point-supervised Brain Tumor Segmentation with Box-prompted Medical Segment Anything Model. 2011 IEEE Nuclear Science Symposium Conference Record 2024, 00: 1-1. PMID: 39445308, PMCID: PMC11497479, DOI: 10.1109/nss/mic/rtsd57108.2024.10656071.Peer-Reviewed Original ResearchSparsity Constrained Linear Tangent Space Alignment Model (LTSA) for 3d Cardiac Extracellular Volume Mapping
Mounime I, Lee W, Marin T, Han P, Djebra Y, Eslahi S, Gori P, Angelini E, Fakhri G, Ma C. Sparsity Constrained Linear Tangent Space Alignment Model (LTSA) for 3d Cardiac Extracellular Volume Mapping. 2024, 00: 1-5. DOI: 10.1109/isbi56570.2024.10635692.Peer-Reviewed Original ResearchT)-space dataExtracellular volume mappingIntrinsic low-dimensional manifold structureCardiac tissue propertiesImproved image reconstructionLow-dimensional manifold structureExtracellular volumeFast MRI methodSparsity constraintModel-based methodsSuperior performanceSpace alignmentT1 mappingManifold structureImage reconstructionT)-spacePost-contrast T1 mappingTissue propertiesFree breathingConcentration of contrast agentLongitudinal relaxation timeAlignment modelDynamic MR imagingSparsityAlignment matrixA deep learning-based approach to nuisance signal removal from MRSI data aqcuired without suppression
Lee W, Zhuo Y, Marin T, Han P, Chi D, Fakhri G, Ma C. A deep learning-based approach to nuisance signal removal from MRSI data aqcuired without suppression. Proceedings Of The International Society For Magnetic Resonance In Medicine ... Scientific Meeting And Exhibition. 2024 DOI: 10.58530/2024/0259.Peer-Reviewed Original ResearchDeep learning-based methodsLearning-based methodsU-Net structureSignal removalIn vivo MRSI dataNeural networkU-NetMRSI dataImage reconstructionSuperior performanceData processingRobust performanceHankel matrixNetworkNuisance signalsConventional methodsPerformanceMRSI signalsSignalMethodRemove nuisance signalsRemovalHankel
2023
Fine-Tuning Network in Federated Learning for Personalized Skin Diagnosis
Lee K, Lee H, Cavalcanti T, Kim S, El Fakhri G, Lee D, Woo J, Hwang J. Fine-Tuning Network in Federated Learning for Personalized Skin Diagnosis. Lecture Notes In Computer Science 2023, 14222: 378-388. DOI: 10.1007/978-3-031-43898-1_37.Peer-Reviewed Original ResearchFederated learningSkin disease diagnosisMobile devicesState-of-the-art approachesUtilization of mobile devicesFine-tuning networkState-of-the-artFine-tuning methodAbstract Federated learningDeep learning networkField of medical diagnosisDeep networksLearning networkAdaptive mannerModified GADisease diagnosisGenetic algorithmSuperior performanceMedical diagnosisArchitectural designClinical datasetsExperimental resultsNetworkModel designPersonality diversity
2022
Joint spectral quantification of MR spectroscopic imaging using linear tangent space alignment‐based manifold learning
Ma C, Han P, Zhuo Y, Djebra Y, Marin T, Fakhri G. Joint spectral quantification of MR spectroscopic imaging using linear tangent space alignment‐based manifold learning. Magnetic Resonance In Medicine 2022, 89: 1297-1313. PMID: 36404676, PMCID: PMC9892363, DOI: 10.1002/mrm.29526.Peer-Reviewed Original ResearchConceptsSubspace-based methodsManifold learningIntrinsic low-dimensional structureGlobal coordinationLearning-based methodsNumerical simulation dataSpatial smoothness constraintSparsity constraintSpace alignmentSubspace modelSmoothness constraintSuperior performanceRoot mean square errorLinear transformationMechanical simulationsLow-dimensionalSquare errorSubspaceExperimental dataSpectroscopic imagingQuantum mechanical simulationsCoordinate alignmentMR spectroscopic imagingSpectral quantificationSimulated dataStructure-aware unsupervised tagged-to-cine MRI synthesis with self disentanglement
Liu X, Xing F, Prince J, Stone M, El Fakhri G, Woo J. Structure-aware unsupervised tagged-to-cine MRI synthesis with self disentanglement. Proceedings Of SPIE--the International Society For Optical Engineering 2022, 12032: 120321q-120321q-7. PMID: 36203947, PMCID: PMC9533681, DOI: 10.1117/12.2610655.Peer-Reviewed Original ResearchStructure feature extractorImage style transferSelf-training schemeStyle transferFeature extractorAdversarial gameSynthesized imagesInformation w.Superior performanceStructural consistencyTask-specificCycle reconstructionStructural encodingImagesDiscoGANCycleGANEncodingExtractorExtraction stepNetworkSchemeGameInput
2021
Domain Generalization under Conditional and Label Shifts via Variational Bayesian Inference
Liu X, Hu B, Jin L, Han X, Xing F, Ouyang J, Lu J, El Fakhri G, Woo J. Domain Generalization under Conditional and Label Shifts via Variational Bayesian Inference. 2021, 881-887. DOI: 10.24963/ijcai.2021/122.Peer-Reviewed Original ResearchDomain generalizationDomain-invariant feature learningVariational Bayesian inference frameworkLabel shiftCross-domain accuracyLabeled source domainVariational Bayesian inferenceFeature learningBayesian inference frameworkLatent spaceSource domainTarget domainDistribution matchingTransfer knowledgeInference frameworkSuperior performanceP(x|yBayesian inferenceLabelingDomainFrameworkGeneralizationBenchmarksPosterior alignmentLearningDual-Cycle Constrained Bijective Vae-Gan For Tagged-To-Cine Magnetic Resonance Image Synthesis
Liu X, Xing F, Prince J, Carass A, Stone M, Fakhri G, Woo J. Dual-Cycle Constrained Bijective Vae-Gan For Tagged-To-Cine Magnetic Resonance Image Synthesis. 2011 IEEE International Symposium On Biomedical Imaging: From Nano To Macro 2021, 00: 1448-1452. PMID: 34707796, PMCID: PMC8547333, DOI: 10.1109/isbi48211.2021.9433852.Peer-Reviewed Original ResearchMR image synthesisAdversarial trainingImage synthesisVAE-GANMR imagingTagged MR imagesCine MR imagingMoving organsSuperior performanceMagnetic resonance imagingAcquisition timeCycle reconstructionHealthy subjectsResonance imagingAnatomical resolutionComparison methodMotion analysisTagged magnetic resonance imagingTissue segmentationImagesScanning sessionA Unified Conditional Disentanglement Framework For Multimodal Brain Mr Image Translation
Liu X, Xing F, Fakhri G, Woo J. A Unified Conditional Disentanglement Framework For Multimodal Brain Mr Image Translation. 2011 IEEE International Symposium On Biomedical Imaging: From Nano To Macro 2021, 00: 10-14. PMID: 34567419, PMCID: PMC8460116, DOI: 10.1109/isbi48211.2021.9433897.Peer-Reviewed Original Research
2020
Severity-Aware Semantic Segmentation with Reinforced Wasserstein Training
Liu X, Ji W, You J, Fakhri G, Woo J. Severity-Aware Semantic Segmentation with Reinforced Wasserstein Training. 2020, 00: 12563-12572. DOI: 10.1109/cvpr42600.2020.01258.Peer-Reviewed Original ResearchSemantic segmentationCARLA simulatorCross-entropyGround distance matrixWasserstein training frameworkAlternating optimization schemeCityscapes datasetDNN architecturesCE lossTraining frameworkSemantic classesGround metricInter-class correlationAutonomous vehiclesSuperior performanceOptimization schemeDNNCARLADistance matrixSurgery systemCamVidDeepLabSimulationCityscapesPixel
2019
Arterial spin labeling MR image denoising and reconstruction using unsupervised deep learning
Gong K, Han P, Fakhri G, Ma C, Li Q. Arterial spin labeling MR image denoising and reconstruction using unsupervised deep learning. NMR In Biomedicine 2019, 35: e4224. PMID: 31865615, PMCID: PMC7306418, DOI: 10.1002/nbm.4224.Peer-Reviewed Original ResearchConceptsSignal-to-noise ratioImage denoisingReconstruction frameworkDeep learning-based image denoisingDeep learning-based denoisersMR image denoisingLearning-based denoisingLow signal-to-noise ratioK-space dataNoisy imagesTraining labelsTraining pairsNetwork inputNeural networkDenoisingIn vivo experiment dataSuperior performanceImaging speedReconstruction processImage qualityLong imaging timesNetworkFrameworkImagesSpatial resolution
2018
Super-Resolution PET Using A Very Deep Convolutional Neural Network
Song T, Chowdhury S, Kim K, Gong K, Fakhri G, Li Q, Dutta J. Super-Resolution PET Using A Very Deep Convolutional Neural Network. 2018, 00: 1-2. DOI: 10.1109/nssmic.2018.8824683.Peer-Reviewed Original ResearchConvolutional neural networkNeural networkSuper-resolution convolutional neural networkDeep convolutional neural networkImage deblurring approachesInput image patchesBlur kernelResolution recovery techniquesSpatial location informationDeblurring approachDeblurring processImage patchesQuantitative accuracy of PETLocation informationSpatially-varying natureSuperior performanceNetworkRecovery techniquesDigital phantomDeblurringBrainWebInformationPartial volume effectsDeepBlur
2014
Numerical Observer for Objective Assessment on Carotid Plaque Using Spectral CT
Lorsakul A, Fakhri G, Ouyang J, Worstell W, Rakvongthai Y, Laine A, Li Q. Numerical Observer for Objective Assessment on Carotid Plaque Using Spectral CT. 2014, 1-4. DOI: 10.1109/nssmic.2014.7430906.Peer-Reviewed Original ResearchMulti-energy CTNumerical observationsCarotid plaquesMatched filterCT systemDigital anthropomorphic phantomHotelling observerDual-energyPlaque featuresImage binConventional CT imagesConventional CT systemsChannelized Hotelling observerAnthropomorphic phantomClassification performanceCT imagesSpectral CTCalcified plaqueClinical classification taskProcessing stepsSimulated imagesPlaqueSuperior performance