Support vector machines for temporal classification of block design fMRI data

• Published in: NeuroImage (Orlando, Fla.) Vol. 26; no. 2; pp. 317 - 329
• Main Authors: LaConte, Stephen, Strother, Stephen, Cherkassky, Vladimir, Anderson, Jon, Hu, Xiaoping
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.06.2005; Elsevier Limited
• Subjects: Algorithms; Artificial Intelligence; Brain - anatomy & histology; Canonical variates analysis; Classification; Computer Graphics; Data analysis; Experiments; Functional magnetic resonance imaging; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Medical imaging; Models, Statistical; NMR; Nuclear magnetic resonance; Studies; Support vector machine; Functional magnetic resonance imaging; Canonical variates analysis; Support vector machine
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2005.01.048

This paper treats support vector machine (SVM) classification applied to block design fMRI, extending our previous work with linear discriminant analysis [LaConte, S., Anderson, J., Muley, S., Ashe, J., Frutiger, S., Rehm, K., Hansen, L.K., Yacoub, E., Hu, X., Rottenberg, D., Strother S., 2003a. The evaluation of preprocessing choices in single-subject BOLD fMRI using NPAIRS performance metrics. NeuroImage 18, 10–27; Strother, S.C., Anderson, J., Hansen, L.K., Kjems, U., Kustra, R., Siditis, J., Frutiger, S., Muley, S., LaConte, S., Rottenberg, D. 2002. The quantitative evaluation of functional neuroimaging experiments: the NPAIRS data analysis framework. NeuroImage 15, 747–771]. We compare SVM to canonical variates analysis (CVA) by examining the relative sensitivity of each method to ten combinations of preprocessing choices consisting of spatial smoothing, temporal detrending, and motion correction. Important to the discussion are the issues of classification performance, model interpretation, and validation in the context of fMRI. As the SVM has many unique properties, we examine the interpretation of support vector models with respect to neuroimaging data. We propose four methods for extracting activation maps from SVM models, and we examine one of these in detail. For both CVA and SVM, we have classified individual time samples of whole brain data, with TRs of roughly 4 s, thirty slices, and nearly 30,000 brain voxels, with no averaging of scans or prior feature selection.