Towards artificial intelligence in mental health: a comprehensive survey on the detection of schizophrenia

• Published in: Multimedia tools and applications Vol. 82; no. 13; pp. 20343 - 20405
• Main Authors: Tyagi, Ashima, Singh, Vibhav Prakash, Gore, Manoj Madhava
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.05.2023; Springer Nature B.V
• Subjects: Artificial intelligence; Autism; Cognition; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Deep learning; Diagnosis; Feature extraction; Machine learning; Medical imaging; Mental disorders; Mental health; Multimedia Information Systems; Neuroimaging; Schizophrenia; Special Purpose and Application-Based Systems; Track 2: Medical Applications of Multimedia; Neuroimaging; Deep learning; Schizophrenia; MRI; EEG; Machine learning
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-022-13809-9

Computer Aided Diagnosis systems assist radiologists and doctors in the early diagnosis of mental disorders such as Alzheimer’s, bipolar disorder, depression, autism, dementia, and schizophrenia using neuroimaging. Advancements in Artificial Intelligence (AI) have leveraged neuroimaging research to unfold numerous techniques for analyzing and interpreting thousands of scans in order to detect and classify various mental illnesses. Schizophrenia is a long-standing psychiatric disorder affecting millions of people worldwide. It causes hallucinations, delusions, and defacement in thinking, behavior, and cognition. Machine Learning and Deep Learning are the subsets of AI which are used for the detection and diagnosis of schizophrenia by gathering insights from different types of modalities. This paper work examines several methods of AI used for the automated diagnosis of schizophrenia using three primary modalities- EEG, structural MRI, and functional MRI. This paper explores different datasets available for schizophrenia along with the techniques and software used to pre-process the EEG and MR images. Further this paper focuses on the different feature extraction and selection techniques to retrieve an appropriate set of features along with the brief overview of machine learning & deep learning approaches. We have also reviewed numerous studies on the prognosis of schizophrenia and presented an exhaustive analysis of the machine learning and deep learning techniques used across EEG and MRI.