The Detection and Classification of Schizophrenia using DL and ML Methods: An Overview of the Recent Works

Schizophrenia (SZ) is a psychotic disorder in which people face delusion, hallucinations, and various behavioral problems. It is tough to identify a patient with this disease by only observing the external physical features. Therefore, advanced technology should be introduced to identify and classify the problem. Recently, Machine Learning (ML) and Deep learning (DL) methods have manifested a great improvement in the field of detection and classification of this disease. Magnetic Resonance Imaging (MRI) and Electroencephalography (EEG) data could be effectively classified using these methods. This review paper includes the evaluations of the ML and DL methods, datasets, limitations, and distinctions of the models, a description of the models, and future scope in this field. The comparative study between used models, their effectiveness, and future scopes will help the researchers to explore this field of research. Researchers can have knowledge about the pros and cons of the methods used in state-of-the-art which will help them to improve the existing methods and also establish a novel practically useable model to ensure an early detection of the disease. This paper would help as a foundation for future research directions in this sector.