Detecting Schizophrenia from Healthy control using Machine Learning and EEG Signals

A brain condition called schizophrenia has a significant impact on people's thinking, behaviors, and lives on a global state. The prevalence of the condition rises along with the population, and those who suffer from schizophrenia lose their capacity for thought and behavior with time. All these issues arise because it is challenging to identify schizophrenia patients. As a result, proper care and quality patient treatment require an accurate and quick diagnosis procedure. This article aims to improve the accuracy of. detection by using EEG signals and Machine Learning algorithms. It is tested using the Resting state EEG dataset, which demonstrates a high level of capability in distinguishing between schizophrenia patients and healthy controls. The EEG data is converted to different forms, where the given machine learning techniques extract the EEG signals to identify patients with schizophrenia and healthy control. We are utilizing the 28-file public dataset in this study. In the file, there are 14 healthy individuals as well as 14 SCZ sufferers. In accordance with the findings, (1) the decision tree successfully distinguished SCZ patients from healthy individuals using EEG signals with an accuracy of 89.12%, and (2) logistic regression was also used to successfully identify the affected patient using EEG signals with an accuracy of 69.78%. (3) The SCZ patient was correctly identified by SVM with an accuracy of 65.51 %. Our research demonstrates the potential for machine learning to play a significant role in aiding in the diagnosis of schizophrenia.