Orthogonal Transform Coding Theory for Schizophrenia Disorder Detection

Abstract

This article presents a method for detection of cognitive defects based on eye tracking and saliency maps. The main goal of our paper is to augment the original approach using saliency maps and correlation matrices by introducing orthogonal transforms, namely–DHYT, DST, DCT2 and DCT3. It was shown that combination of various orthogonal transforms provided different results, which are discussed in the article. Experimental data were measured using eye tracker tool for a group of healthy (22) and diagnosed (22) individuals exposed to the Rorschach inkblot test. On average the measured accuracies reached scores over 70% which forms a solid basis for automated prescreening deployment.