2019
Combining multiple connectomes improves predictive modeling of phenotypic measures
Gao S, Greene AS, Constable RT, Scheinost D. Combining multiple connectomes improves predictive modeling of phenotypic measures. NeuroImage 2019, 201: 116038. PMID: 31336188, PMCID: PMC6765422, DOI: 10.1016/j.neuroimage.2019.116038.Peer-Reviewed Original ResearchConceptsMultiple connectomesLarge open-source datasetOpen-source datasetNovel prediction frameworkPredictive modelingSingle predictive modelPredictive modelArt algorithmsPrediction frameworkMultiple tasksPredictive model approachPrincipled waySpecific algorithmsFunctional connectivity matricesConnectivity matrixDifferent tasksPrediction performanceConnectome-based predictive modelingHuman Connectome ProjectTaskSuperior performanceAlgorithmComplementary informationNaïve extensionsConnectome Project
2018
Combining Multiple Connectomes via Canonical Correlation Analysis Improves Predictive Models
Gao S, Greene A, Todd Constable R, Scheinost D. Combining Multiple Connectomes via Canonical Correlation Analysis Improves Predictive Models. Lecture Notes In Computer Science 2018, 11072: 349-356. DOI: 10.1007/978-3-030-00931-1_40.Peer-Reviewed Original ResearchTask conditionsDifferent cognitive tasksMultiple task conditionsDifferent task conditionsConnectivity dataDifferent cognitive conditionsFunctional connectivity dataComputational modelHuman Connectome ProjectPrediction of behaviorCognitive tasksIndividual differencesBehavioral measuresBehavioral predictionsCognitive conditionsMultiple connectomesSingle taskFunctional connectivityConnectome ProjectDifferent tasksComplementary informationMultiple tasksTaskPrincipled methodCanonical correlation analysis