Abstract
Depression is one of the most prevalent mental disorders with significant disability and mortality rates, which severely impairs the health and social function of individuals. To explore more efficient and objective diagnosis and treatment methods, the application of artificial intelligence technology in depression has gained increasing attention in recent years, providing a new tool for clinical practice and achieving preliminary successes. The goal of this study was to explore the research hotspots and frontiers of AI in the field of depression, providing reference for future research. Methods: We used literature related to AI in the field of depression retrieved from Web of Science as the data source. In combination with manual screening and information integration methods, CiteSpace software was used to make a statistical analysis of literature distribution, authors, institutions, keywords. Results: A total of 500 English articles were included, and the number of English articles increased significantly from 2019 to 2022. The authors were mainly new research groups in recent years. The pattern of institutional collaboration was relatively simple, and there were few highly cited articles. The research methods mainly involved data mining, machine learning, deep learning, and neural networks, while the application areas mainly focusing on the use of AI technology for predicting depression onset, early recognition, auxiliary diagnosis, and treatment planning. Conclusion: The research hotspots of AI in the field of depression are centered on improving depression intelligent methods, intelligent diagnosis, and intelligent prediction. The focus is more inclined towards intelligent prediction and diagnosis research. Research teams and institutions should strengthen cooperation among experts in different research fields to achieve multi-disciplinary integration.

This publication has 12 references indexed in Scilit: