Pattern Identification and Acupuncture Prescriptions Based on Real-World Data Using Artificial Intelligence

Abstract

In traditional East Asian medicine, acupuncture practitioners gather clinical data, identify patterns, and choose appropriate acupoints. The pattern identification procedure is crucial for clinical diagnosis and acupuncture treatment. Understanding the pattern identification process, i.e. gathering and synthesizing clinical information from patient signs and symptoms, is crucial for characterizing the complicated relationships between symptoms and acupoints. Here, we briefly overview recent studies describing the use of a bodily sensation map to identify spatial patterns of “acupoint indications”, an artificial neural network model to characterize the rules connecting symptoms with acupoints, and medical data extrapolated from case reports to reveal associations between diagnoses and acupoint prescriptions. We also propose a method based on pattern identification to optimize acupoint selection for treatment. Artificial intelligence has substantially advanced traditional East Asian medicine by facilitating decision-making and aids understanding of clinical decision-making as it relates to acupuncture treatment.