biostudies-literaturePettersson-Yeo WWellcome Trust189https://www.ebi.ac.uk/biostudies/studies/S-EPMC4097812biostudies-literatureUnknown8In the pursuit of clinical utility, neuroimaging researchers of psychiatric and neurological illness are increasingly using analyses, such as support vector machine, that allow inference at the single-subject level. Recent studies employing single-modality data, however, suggest that classification accuracies must be improved for such utility to be realized. One possible solution is to integrate different data types to provide a single combined output classification; either by generating a single decision function based on an integrated kernel matrix, or, by creating an ensemble of multiple single modality classifiers and integrating their predictions. Here, we describe four integrative approaches: (1) an un-weighted sum of kernels, (2) multi-kernel learning, (3) prediction averaging, and Frontiers in neuroscienceAn empirical comparison of different approaches for combining multimodal neuroimaging data with support vector machine.PMC4097812WT088641/Z/09/ZWT085390/Z/08/ZMcGuire PPettersson-Yeo WMechelli ACatani MMarquand AFBenetti SAllen PJoules RWilliams SCfalseAn empirical comparison of different approaches for combining multimodal neuroimaging data with support vector machine.In the pursuit of clinical utility, neuroimaging researchers of psychiatric and neurological illness are increasingly using analyses, such as support vector machine, that allow inference at the single-subject level. Recent studies employing single-modality data, however, suggest that classification accuracies must be improved for such utility to be realized. One possible solution is to integrate different data types to provide a single combined output classification; either by generating a single decision function based on an integrated kernel matrix, or, by creating an ensemble of multiple single modality classifiers and integrating their predictions. Here, we describe four integrative approaches: (1) an un-weighted sum of kernels, (2) multi-kernel learning, (3) prediction averaging, and 2014-01-01T00:00:00Z20142025-04-22T00:59:50.653Z2019-03-27T01:32:04ZS-EPMC40978122507686810.3389/fnins.2014.00189