A systematic review of artificial intelligence for mental rehabilitation

Background and aims: As artificial intelligence (AI) continues to advance, its role in medicine,including psychology, cannot be ignored. This article provides an overview of the applicationsof AI in mental health and rehabilitation of mental disorders, as well as the challenges and opportunitiesit presents. It also briefly discusses the ethical implications of using AI in psychiatry,psychology, and psychotherapy.Method: The investigation searched four medical databases (PubMed, Scopus, ScienceDirect,and PsycINFO) using specific keywords such as artificial intelligence, machine learning, datamining, mental rehabilitation, psychiatry, mental health, and mental disorder. Studies that werenot in English language, case studies, reviews, conference papers, and those that followed anonymizationprocedures were excluded as part of the criteria. No restrictions were placed onpublication dates.Results: A total of ۳۶۹ articles were identified, of which ۶۲ were included in the review that usedmedical databases (EHRs), patients who came to the emergency room, novel monitoring systems,brain imaging data, and social media users to classify mental health illnesses. The literature reviewrevealed a wide range of applications with three main themes: diagnosis, prognosis, treatmentand support, treatment, public health, and research. The studies on AI in patient flow identifiedreadmissions, resource allocation, and limitations as the primary themes. The most commonmental health conditions addressed included depression, schizophrenia, Alzheimer’s disease, andsuicide ideation and attempts. Natural language processing (NLP) and machine learning (ML)techniques used included support vector machines, decision trees, neural networks, latent Dirichletallocation, and clustering.Conclusion: This text discusses the use of machine learning and natural language processing inmental health research. While these techniques have the potential to provide valuable insightsfrom unexplored data sources, ethical concerns need to be addressed before they can be used asclinical tools. Despite this, there are already significant benefits to using ML in mental health,particularly in areas like diagnosis and treatment. However, more research is needed to exploreother potential applications, such as improving patient flow across different specialties. As AItechniques continue to improve, it may be possible to redefine mental illnesses, identify them atan earlier stage, and personalize treatments based on individual characteristics. Nonetheless, cautionmust be taken not to over-interpret initial results or lose sight of the importance of bridgingthe gap between AI research and clinical care.

کلیدواژه ها:

artificial intelligence ، mental rehabilitation ، natural language processing ، psychiatry ، machine learning