Preclinical Diagnosis of Schizophrenia using EEGTime-Frequency Analysis During Motor Function and verbal Fluency Tests

Abstract Schizophrenia is a severe mental illness that impairs the way a person perceives reality. It causes a variety of issues related to behavior, emotions, and thinking (cognition). Patients experience auditory hallucinations, delusion, and sleep deprivation. Although the Diagnostic and Statistical Manual (DSM) IV version helps in diagnosis, the lack of clinical tools available leads to delayed diagnosis. To ease preclinical diagnosis, this study conducted two tests, namely Motor Function Test (MFT) and the verbal Fluency Test (VFT). The neural activity of the brain during these two tests is captured by an Electroencephalogram(EEG). However, EEG signals are not widely used for clinical analysis, as the signals are blurred during acquisition, random, and shortcoming of the signal processing algorithm. Hence, in this paper, we propose a Preclinical Diagnosis of Schizophrenia using Multi-SynchroSqueezing Transform (MSST) (PDS-M) to analyze the blurred EEG signal with Time-frequency analysis.PDS-M produces a sharper signal and achieves perfect signal reconstruction. The mono component of MSST in each mode is termed intermediate frequency and it is used as the feature (such as lack of emotion, hallucinations, or disorganized thinking)for discriminating schizophrenic subjects from normal ones. The results show that the MFT is better than the VFT for preclinical study and this study also suggests that MSST is more suitable for time-frequency analysis of EEG signal since all the three modes multi-components are different from normal having a low p-value of p=0, p=0, and p=0.008, respectively, where p is the probability of data having occurred under the null hypothesis of a statistical test.