Abstract
OBJECTIVE: This Systematic Review aims to provide a comprehensive analysis of the current state of anxiety disorder detection methods using Artificial Intelligence (AI), focusing on their accuracy and the scope of research. This review is tailored for researchers, clinicians, and technology developers seeking to understand the advancements in AI-driven mental health diagnostics. METHODOLOGY: A Systematic Review was conducted following the PRISMA Statement guidelines, utilizing databases such as IEEE Xplore, PubMed, ScienceDirect, and SpringerLink. The review included studies focusing on the diagnosis of anxiety disorders using quantitative data and AI techniques, excluding those solely focused on depression or lacking experimental datasets. RESULTS: A total of 119 studies were analyzed, revealing the application of Machine Learning and Deep Learning techniques in detecting anxiety disorders from diverse data sources, including self-reports, physiological data, and social network data. The findings indicate that AI-driven methods demonstrate higher accuracy compared to traditional anxiety disorder detection tests, providing valuable insights for clinicians and researchers exploring improved diagnostic tools. CONCLUSIONS: This review highlights the critical role of AI in optimizing the detection and treatment of anxiety disorders. It offers a current and detailed overview of advancements in this field, making it a key resource for researchers, healthcare professionals, and technology developers aiming to integrate AI into mental health practices. The synthesis of findings provides a clear understanding of the current landscape and potential future directions in AI-based anxiety detection. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420251026205, identifier CRD420251026205.