A bibliometric analysis of artificial intelligence applications in anxiety disorder diagnosis and treatment

Abstract Anxiety disorders are prevalent mental health conditions characterized by intense worry, fear, and apprehension, which negatively impact individuals’ quality of life and functional capacity. These disorders encompass various subtypes, including separation anxiety, selective mutism, specific phobia, social anxiety, panic disorder, agoraphobia, and generalized anxiety disorder. If left untreated, anxiety disorders can adversely affect individuals’ social, academic, and occupational functioning and may lead to the development of additional psychological problems. In recent years, artificial intelligence based technologies have offered innovative solutions in the diagnosis and treatment of anxiety disorders, contributing to more objective and effective diagnostic processes. This study aims to examine scientific publications that employ Artificial intelligence approaches in the diagnosis and treatment of anxiety disorders using bibliometric analysis methods. The bibliometric analysis, which included 2491 studies retrieved from the Scopus and Web of Science databases, reveals that the field exhibits multidimensional and interdisciplinary development both theoretically and practically. Furthermore, scientific production in this domain is characterized by a collaborative and dynamic structure. Specifically, the analysis reveals that the United States and China are the leading countries in scientific production, while the most robust international collaborations occur among the USA, China, and the United Kingdom. The thematic focus of the literature is largely driven by keywords such as machine learning, depression, anxiety, mental health, and deep learning. The findings indicate that artificial intelligence and machine learning-based approaches are increasingly being utilized in the diagnosis, treatment, and prediction processes of anxiety disorders. Moreover, research topics in this field are evolving over time toward more specific, clinically oriented, and practice-based directions. These results suggest that artificial intelligence applications will continue to represent a significant area of research and practice within the domain of psychiatric disorders.