Bulgarian Journal of Psychiatry, 2025; 10(1):32-39
The use of artificial intelligence and applications to identify and treat mental disorders
Liana Spytska
Department of Psychology and Pedagogy, Kyiv International University, Kyiv, Ukraine
Abstract. Introduction. Mental illness is a serious burden on the modern world. The number of people with various psychiatric conditions was over 1 billion as of 2023 alone, and that number is on a permanent rise. The use of artificial intelligence (AI) in this regard will make mental health care more accessible. The purpose of this study was to investigate the role of using АI chatbots with people with mental disorders. Materials and methods. This study examined 38 sources of information in databases. Furthermore, 40 patients who agreed to work with the chatbots under doctor’s supervision were also selected, and the effеcts of the intervention were assessed. Results. AI is arguably very promising to use for people with mental disorders, but its usefulness varies depending on the diagnosis and symptoms. Patients diagnosed with paranoid schizophrenia who have predominantly positive symptoms in the form of hallucinations and delusions did not show sufficient treatment improvement – some of them, on the contrary, had experienced deterioration, especially with delusions of machines overtaking the world. However, positive trend was observed in patients with depression and anxiety disorder: they felt less lonely, because they could contact the chatbot whenever they wanted and needed. Despite the positive trend, there was no great efficacy as expected. It was suggested that this is due to the small number of patients. A reduction in suicidal thoughts was also found. Based on the results of this study, the use of AI may be reasonable in psychiatry, but preferably at least partially supervised by a doctor. Conclusion. The data from this study shows the effiсacy of chatbots use which appears to be limited. Therefore, studies should be expanded with greater numbers of patients to obtain reliable results.
Key words: psychiatry, depression, schizophrenia, chatbots, machine algorithms
Full text here.