Schizophrenia Patients Exhibit Aberrant Phase-Space Embedding of CEN/DMN Interaction

A proper method for distinguishing Schizophrenia (SZ) patients from people belonging to normal controls is invaluable for diagnostics. Standard techniques of identifying SZ patients include a variety of clustering-based methods. For example, those based on independent component analysis (ICA). However, they typically focus on average connectivity. In this work, we propose a novel method for SZ patient identification based on the products of ICA of default mode network (DMN) and cognitive enhancement therapy (CET). Numerical experiments show that the proposed method achieves statistically significant results for differentiating the two groups of subjects.