Multiclass classification for brain state prediction in fMRI

Abstract

The multi-voxel pattern analysis technique is applied to fMRI data for classification of high-level brain function using pattern information distributed over multiple voxels. The fMRI signal analysis requires multiclass classification rather than binary classification. The paper proposes a multiclass classifier for fMRI analysis using pairwise classifier ensemble. Each pairwise classifier consists of multiple sub-classifiers optimized by a customized searchlight analysis to utilize an adaptive feature set for each class-pair. The results of multiple pairwise classifiers are combined to estimate the classification result. Simulated and real fMRI data are used to verify the proposed method. Intra- and inter-subject analyses are performed to compare the proposed method with several well-known classifiers, including single and ensemble classifiers. The comparison results show that the proposed method can be generally applied to multiclass classification in both simulations and real fMRI analyses.