Self-supervised multimodal learning for group inferences from MRI data: Discovering disorder-relevant brain regions and multimodal links
Fedorov A, Geenjaar E, Wu L, Sylvain T, DeRamus T, Luck M, Misiura M, Mittapalle G, Hjelm R, Plis S, Calhoun V. Self-supervised multimodal learning for group inferences from MRI data: Discovering disorder-relevant brain regions and multimodal links. NeuroImage 2023, 285: 120485. PMID: 38110045, PMCID: PMC10872501, DOI: 10.1016/j.neuroimage.2023.120485.Peer-Reviewed Original ResearchConceptsBrain regionsMultimodal neuroimaging dataNeuroimaging dataBrain disordersComplex brain disordersMRI dataNeuroimaging researchGroup inferencesDeep InfoMaxSupervised modelsDiagnostic labelsDisordersBrainState-of-the-art unsupervised methodsAlzheimer's phenotypeNovel self-supervised frameworkSelf-supervised frameworkSelf-supervised methodologyCanonical correlation analysisSelf-supervised representationsState-of-the-artDeep learning approachSingle-modal dataMultimode linksComplex brainsAny-Way Independent Component Analysis with Reference
Duan K, Silva R, Liu J, Agcaoglu O, Calhoun V. Any-Way Independent Component Analysis with Reference. 2023, 00: 1-4. DOI: 10.1109/isbi53787.2023.10230369.Peer-Reviewed Original ResearchCross-modal correlationIndependent component analysisMultiset canonical correlation analysisOptimal global solutionMultimodal fusionNoisy conditionsOrthogonality requirementCanonical correlation analysisIndependence of sourcesJoint independent component analysisSimulation resultsComponent analysisImproved accuracyComponent matricesGlobal solutionICAMultisetsMultimodal patternsOrthogonalityMultimodal Subspace Independent Vector Analysis Better Captures Hidden Relationships in Multimodal Neuroimaging Data
Li X, Adali T, Silva R, Calhoun V. Multimodal Subspace Independent Vector Analysis Better Captures Hidden Relationships in Multimodal Neuroimaging Data. 2023, 00: 1-5. DOI: 10.1109/isbi53787.2023.10230605.Peer-Reviewed Original ResearchSubspace structureIndependent vector analysisSynthetic datasetsMultimodal neuroimaging datasetUnimodal analysisData modalitiesHidden relationshipsCanonical correlation analysisIncorrect onesNeuroimaging datasetsSubspaceLatent sourcesDatasetNeuroimaging modalitiesDataPhenotypic measurementsCorrelation analysisData-driven multimodal fusion: approaches and applications in psychiatric research
Sui J, Zhi D, Calhoun V. Data-driven multimodal fusion: approaches and applications in psychiatric research. Psychoradiology 2023, 3: kkad026. PMID: 38143530, PMCID: PMC10734907, DOI: 10.1093/psyrad/kkad026.Peer-Reviewed Original ResearchMultimodal fusionFusion methodBig dataMulti-modal fusion methodMulti-modal fusion techniquesImage fusion applicationsEra of big dataMultimodal fusion approachDeep learning approachExtract meaningful insightsIndependent component analysisFusion approachHidden patternsLearning approachFusion techniqueCanonical correlation analysisPrior informationMultiple modalitiesComplex psychiatric disorderN-wayMeaningful insights