Incorporating AI Into Military Behavioral Health: A Narrative Review

作者：Ann D McConnon, Airyn J Nash, John Ray Roberts, Shmuel Z Juni, Ashley Derenbecker, Patrice Shanahan, Andrew J. Waters · 发表于：Military Medicine · 年份：2025 · DOI：10.1093/milmed/usaf162 · 被引用次数：5 · 研究领域：Digital Mental Health Interventions、Mental Health via Writing、Traumatic Brain Injury Research

INTRODUCTION: Concerns regarding suicide rates and declining mental health among service members highlight the need for impactful approaches to address behavioral health needs of U.S. military populations and to improve force readiness. Research in civilian populations has revealed that artificial intelligence and machine learning (AI/ML) have the promise to advance behavioral health care in the following 6 domains: Education and Training, Screening and Assessment, Diagnosis, Treatment, Prognosis, and Clinical Documentation and Administrative Tasks. MATERIALS AND METHODS: We conducted a narrative review of research conducted in U.S. military populations, published between 2019 and 2024, that involved AI/ML in behavioral health. Studies were extracted from Embase, PubMed, PsycInfo, and Defense Technical Information Center. Nine studies were considered appropriate for the review. RESULTS: Compared to research in civilian populations, there has been much less research in U.S. military populations regarding the use of AI/ML in behavioral health. The studies selected using ML have shown promise for screening and assessment, such as predicting negative mental health outcomes in military populations. ML has also been applied to diagnosis as well as prognosis, with initial positive results. More research is needed to validate the results of the studies reviewed. CONCLUSIONS: There is potential for AI/ML to be applied more extensively to military behavioral health, including education...