Application of Neural Networks and Machine Learning Models in the Diagnosis, Classification and Individual Treatment of Schizophrenia

Abstract

Introduction. Schizophrenia is a chronic, severe, and progressive mental disorder whose course is influenced by genetic, epigenetic, and environmental factors. Symptoms typically appear between the ages of 16 and 30 and can be divided into three categories: positive, negative and cognitive symptoms. In recent years, the use of artificial intelligence techniques in psychiatry has increased. Numerous studies demonstrate the effectiveness of methods such as deep learning in the diagnosis and treatment of psychiatric disorders.

Aim of the study. To summarize the current state of knowledge regarding the application of neural networks and machine learning models in the diagnosis, classification and personalized treatment of schizophrenia.

Methods and materials. A literature review was conducted using publications available in the PubMed and Google Scholar databases from 2019-2024. The most relevant and recent studies were selected using appropriate keywords.

Results. Artificial intelligence techniques enable the development of diagnostic and prognostic models identifying patients with schizophrenia based on MRI and EEG data, as well as audio, video recordings and genetic information. Patient classification is mainly based on EEG and MRI analyses. However, the use of machine learning for treatment personalization remains limited. Ethical issues, data protection, and the nature of the physician-patient relationship in psychiatry remain important challenges.

Conclusions. Artificial intelligence shows potential in the diagnosis, prognosis and treatment of schizophrenia, but its implementation in clinical practice requires further research, methodological improvement and careful consideration of ethical and legal aspects.