2024
Explainable fuzzy clustering framework reveals divergent default mode network connectivity dynamics in schizophrenia
Ellis C, Miller R, Calhoun V. Explainable fuzzy clustering framework reveals divergent default mode network connectivity dynamics in schizophrenia. Frontiers In Psychiatry 2024, 15: 1165424. PMID: 38495909, PMCID: PMC10941842, DOI: 10.3389/fpsyt.2024.1165424.Peer-Reviewed Original ResearchHard clusteringNetwork dynamicsDynamic functional network connectivityFuzzy clustering frameworkExtract several featuresFuzzy clusteringC-meansExplainability approachesExplainability metricsData spaceClustering frameworkK-meansDynamic functional network connectivity stateNetwork connectivityModerate anticorrelationImage dataNetworkState dynamicsAnalysis frameworkConnectivity dynamicsFunctional network connectivityAnticorrelationCentroidFunctional magnetic resonance imaging dataFramework
2023
An Explainable and Robust Deep Learning Approach for Automated Electroencephalography-Based Schizophrenia Diagnosis
Sattiraju A, Ellis C, Miller R, Calhoun V. An Explainable and Robust Deep Learning Approach for Automated Electroencephalography-Based Schizophrenia Diagnosis. 2023, 00: 255-259. DOI: 10.1109/bibe60311.2023.00048.Peer-Reviewed Original ResearchConvolutional neural networkRobust deep learning approachBaseline convolutional neural networkChannel lossDeep learning methodsDeep learning modelsDeep learning approachDecision support roleExplainability approachesClassifier performanceRobust modelNeural networkExplainable modelsLearning methodsLearning approachLearning modelsAutomated diagnosisImplementation environmentEEG dataDiagnosis of SZExplainabilityRaw EEGTest dataRobustnessBiomarkers of SZNovel Approach Explains Spatio-Spectral Interactions In Raw Electroencephalogram Deep Learning Classifiers
Ellis C, Sattiraju A, Miller R, Calhoun V. Novel Approach Explains Spatio-Spectral Interactions In Raw Electroencephalogram Deep Learning Classifiers. 2023, 00: 1-5. DOI: 10.1109/icasspw59220.2023.10193605.Peer-Reviewed Original ResearchNovel methods for elucidating modality importance in multimodal electrophysiology classifiers
Ellis C, Sendi M, Zhang R, Carbajal D, Wang M, Miller R, Calhoun V. Novel methods for elucidating modality importance in multimodal electrophysiology classifiers. Frontiers In Neuroinformatics 2023, 17: 1123376. PMID: 37006636, PMCID: PMC10050434, DOI: 10.3389/fninf.2023.1123376.Peer-Reviewed Original ResearchExplainability approachesExplainability methodsAutomated sleep stage classificationRaw time series dataConvolutional neural networkDeep learning classifierSleep stage classificationNovel methodMultimodal classificationLearning classifiersNeural networkClassifierLocal explanationsGlobal explanationsExplainabilitySubject-level differencesTime series dataAdvancement of personalized medicineGlobal methodClinical classifierClassificationClinical variablesElectrophysiological studiesStage classificationElectrophysiological classification