Abstract
Acupuncture, as a traditional medical therapy, has gained widespread application globally. However, traditional acupoint localization has relied heavily on the personal experience of clinicians. With the advancement of AI, especially deep learning, researchers have begun exploring automated acupoint localization in recent years. This study systematically reviews the literature on acupoint localization in different anatomical regions using traditional algorithms and deep learning to evaluate the effectiveness, limitations, and future directions of these technologies. Drawing insights from the surveyed studies, we anticipate that the challenges outlined in this research will be addressed in the near future. As these challenges are overcome, automated acupoint localization techniques are expected to yield more promising outcomes. We hope our efforts will inspire further exploration of advanced AI methods for acupoint localization.