A data-driven analysis of artificial intelligence applications in depression research: 2020-2025

With the rapid advancement of artificial intelligence (AI) technologies, increasing numbers of researchers have explored the integration of AI into depression research. In this study, we conducted a bibliometric analysis of research related to the application of AI in depression research. Our findings revealed a substantial growth trend in publications since 2020 with China and the United States leading this expansion. Institutions such as the University of Toronto and Sichuan University have played prominent roles as key contributors. Notably, prolific authors have advanced the integration of AI algorithms into areas such as depression diagnosis, symptom classification, and risk prediction. This trend is further supported by our keyword co-occurrence analysis, which reveals a strong positive correlation between terms such as "machine learning" and "prediction", as well as between "feature extraction" and "electroencephalography". Moreover, our analysis suggests that while AI applications in depression research share common methodological frameworks, they also exhibit distinct patterns based on age, gender, and population characteristics. This bibliometric study provides a comprehensive overview of research trends, influential contributors, and emerging directions in the field of AI-assisted depression research, offering valuable insights for future interdisciplinary development.