The use of machine learning as a diagnostic and monitoring tool for bipolar disorder

Abstract

Introduction: Bipolar disorder (BD) is a disorder in which patients experience alternating episodes of depression, mania or hypomania, and sometimes mixed episodes. BD is associated with serious comorbidities and a shorter average life expectancy. Both diagnosis and treatment are long-term processes, often requiring prophylaxis for many years. Early diagnosis and rapid intervention may help extend life expectancy through the use of artificial intelligence.

Aim of the study: The aim of this study is to summarize the current state of knowledge regarding the use of machine learning as a tool supporting diagnosis and monitoring the course of bipolar disorder.

Materials and methods: A review of literature available in the PubMed and Google Scholar databases was conducted. The literature used in the article comes from the years 2019–2024. The following keywords were used: bipolar disorder, machine learning, artificial intelligence, diagnosis and treatment of bipolar disorder.

Results: Artificial intelligence plays an important role in the diagnosis and monitoring of patients with BD, supporting differential diagnosis, prediction of relapses, and personalization of treatment. Analyses based on EEG signals and brain imaging suggest that AI may distinguish individuals with BD from healthy individuals, although more consistent research results are needed. AI in psychiatry also faces limitations such as difficulties in accessing large, high-quality datasets and insufficient physician training.

Conclusions: Recent studies demonstrate that artificial intelligence techniques may be effective in the diagnosis and monitoring of bipolar disorder, although machine learning still requires further improvement.

Keywords: bipolar disorder, machine learning, artificial intelligence, diagnosis of bipolar disorder, treatment of bipolar disorder.