2021
Magnetic resonance parameter mapping using model‐guided self‐supervised deep learning
Liu F, Kijowski R, Fakhri G, Feng L. Magnetic resonance parameter mapping using model‐guided self‐supervised deep learning. Magnetic Resonance In Medicine 2021, 85: 3211-3226. PMID: 33464652, PMCID: PMC9185837, DOI: 10.1002/mrm.28659.Peer-Reviewed Original ResearchConceptsMR parameter mappingSupervised learningReconstruction qualityImaging modelSelf-supervised deep learningStandard supervised learningConventional iterative reconstructionData setsDeep learning purposesSuperior reconstruction qualityImprove reconstruction qualityQuantitative MRI applicationsUndersampled k-spacePresence of noisePhysical modeling constraintsSparsity constraintNetwork trainingReconstruction performanceDeep learningReconstruction frameworkMap extractionImprove image qualitySuppress noiseGround truthUndersampling artifacts
2019
Evaluation of pharmacokinetic modeling strategies for in-vivo quantification of tau with the radiotracer [18F]MK6240 in human subjects
Guehl N, Wooten D, Yokell D, Moon S, Dhaynaut M, Katz S, Moody K, Gharagouzloo C, Kas A, Johnson K, El Fakhri G, Normandin M. Evaluation of pharmacokinetic modeling strategies for in-vivo quantification of tau with the radiotracer [18F]MK6240 in human subjects. European Journal Of Nuclear Medicine And Molecular Imaging 2019, 46: 2099-2111. PMID: 31332496, PMCID: PMC6709592, DOI: 10.1007/s00259-019-04419-z.Peer-Reviewed Original ResearchConceptsReference tissue methodDistribution volume ratioTissue methodIn vivo quantificationPharmacokinetic modeling strategiesArterial plasma input functionMultilinear reference tissue methodsTwo-tissue compartment modelBlood:plasma ratioTissue-to-plasmaPlasma input functionPlasma concentration time courseBlood-based methodMethodsThirty-five subjectsSUV ratioBlood-based analysesData setsArterial input functionPET scansControl subjectsMild cognitive impairmentPlasma ratioRadiometabolite analysisHealthy controlsConcentration time course
2018
Attenuation correction for brain PET imaging using deep neural network based on Dixon and ZTE MR images
Gong K, Yang J, Kim K, Fakhri G, Seo Y, Li Q. Attenuation correction for brain PET imaging using deep neural network based on Dixon and ZTE MR images. Physics In Medicine And Biology 2018, 63: 125011. PMID: 29790857, PMCID: PMC6031313, DOI: 10.1088/1361-6560/aac763.Peer-Reviewed Original ResearchConceptsU-Net structureU-NetModified U-net structureAttenuation correctionDeep neural network methodBrain PET imagingPET attenuationDeep neural networksPatient data setsAttenuation coefficientDixon-based methodNeural network methodData setsConvolution moduleNetwork inputNeural networkDixon MRPET/MR hybrid systemImage reconstructionPET imagingNetwork methodNetworkNetwork approachNetwork structureQuantification errors
2016
Validation of Bayesian analysis of compartmental kinetic models in medical imaging
Sitek A, Li Q, Fakhri G, Alpert N. Validation of Bayesian analysis of compartmental kinetic models in medical imaging. Physica Medica 2016, 32: 1252-1258. PMID: 27692754, PMCID: PMC5720163, DOI: 10.1016/j.ejmp.2016.09.010.Peer-Reviewed Original ResearchConceptsAccurate estimation of uncertaintyComputer simulationsMedical imagesPosterior distributionDistributed noiseTime series of imagesClosed-formSeries of imagesData setsKinetic parametersMarkov chain Monte Carlo methodsPosterior distributions of kinetic parametersNon-linear least squares methodAccurate estimationComputerLeast-squares methodKinetic modelEstimation of kinetic parametersF18-fluorodeoxyglucoseBayesian estimationImagesStatistical inferenceMonte Carlo methodEstimates of uncertaintyInformation
2015
Matched signal detection on graphs: Theory and application to brain imaging data classification
Hu C, Sepulcre J, Johnson K, Fakhri G, Lu Y, Li Q. Matched signal detection on graphs: Theory and application to brain imaging data classification. NeuroImage 2015, 125: 587-600. PMID: 26481679, DOI: 10.1016/j.neuroimage.2015.10.026.Peer-Reviewed Original ResearchConceptsImage data classificationWeighted energy detectorGraph-signalGraph Laplacian eigenvaluesEnergy detectorManifold structureProblem of Alzheimer's diseaseData classificationGraph LaplacianSubspace detectorWeighted graphMSD approachSignal processingSignal detectionIntrinsic structureLaplacian eigenvaluesSubspaceTest statisticsGraphRandom signalsData setsLowest eigenvalueGaussian distributionTraditional methodsEigenvalues
2011
A Nonlocal Averaging Technique for Kinetic Parameter Estimation from Dynamic PET Data
Dutta J, Fakhri G, Leahy R, Li Q. A Nonlocal Averaging Technique for Kinetic Parameter Estimation from Dynamic PET Data. 2011, 3562-3566. DOI: 10.1109/nssmic.2011.6153669.Peer-Reviewed Original ResearchNonlocal meansDenoising schemeGaussian filtering techniquePatlak parametric imagesDenoised time seriesDenoising frameworkClustering stepDistributed natureGaussian filterLocal neighborhoodVoxel-wise estimationHigh noise levelsData setsDenoisingFiltering techniqueSchemeDataLow varianceNoise levelTime activity curvesEstimation of kinetic parametersDynamic PET imagesParameter estimationVoxelImages
2001
Comparative Assessment of Energy-Based Methods of Compensating for Scatter and Lead X-Rays in Ga-67 SPECT Imaging
Moore S, Fakhri G, Maksud P. Comparative Assessment of Energy-Based Methods of Compensating for Scatter and Lead X-Rays in Ga-67 SPECT Imaging. 2001, 4: 2197-2198. DOI: 10.1109/nssmic.2001.1009260.Peer-Reviewed Original ResearchLead X-raysGa-67Energy windowArtificial neural networkGa-67 SPECT imagingSPECT imagesHigh-energy contaminationGa-67 SPECTPoisson noise realizationsActivity estimation taskTumor activity concentrationAnthropomorphic phantomEvaluable tumorsGS methodTumorMean square errorData setsOrgan uptakeProjection imagesLymphoma studiesNeural networkPixel valuesX-raySpherical tumorNoise realizations