Detection of Schizophrenia using FMRI Data via Projection Pursuit

Abstract

Schizophrenia is currently diagnosed based upon symptoms and there is no quantitative, biologically based technique as yet. Classification of individuals into schizophrenia and control groups based on fMRI data is thus of great interest to support psychiatric diagnoses. We applied a novel projection pursuit technique on the default mode component of 70 subjects' fMRI data obtained during an auditory oddball task. The validity of the technique was tested with a leave-one-out method and the detection performance varied between 80% and 90% applying different masks. The findings suggest that the proposed data reduction algorithm is effective in classifying individuals into schizophrenia and control groups and useful as a diagnostic tool.