Tabari Biomedical Student Research Journal

Physical inactivity remains a significant concern from a public health standpoint for women and contributes to the development of chronic diseases at every stage of life. Even though tailored exercise results in improved health outcomes, standard exercise prescriptions do not consider women's physiological differences, varying needs, or features linked to different phases of their lives. Artificial intelligence has recently emerged as a promising way of offering personalized, data-driven exercise programs, but there is still a lack of evidence specific to women.
The review was carried out following PRISMA 2020 guidelines and consisted of searching PubMed/MEDLINE, Scopus, Web of Science, and Google Scholar databases for peer-reviewed studies published between January 2020 and March 2026; ten studies met inclusion criteria and employed machine learning, wearable AI, computer vision, automated coaching systems, and large language models to personalize exercise.
In the studies, those that used AI saw improvements in physical activity, body composition, musculoskeletal function, cardiometabolic health, adherence, and behavioral and mental health. Although most of the randomized trials were of high methodological quality, the pilot studies and case studies were of only moderate or lower quality due to limitations in their design.
AI has the definite potential to make exercise interventions more precise, accessible, scalable, and tailored to women, but the existing evidence is limited due to small sample sizes, varying protocols, and short periods of follow-up; it is therefore necessary to conduct larger, higher-quality trials employing standardized measures and involving longer periods of follow-up before routine clinical use can be recommended.