Artificial Intelligence-Based Intelligent Decision Support and Personalized Intervention for Traditional Chinese Medicine External Therapies

Abstract

Traditional Chinese medicine (TCM) external therapies involve diverse modalities, individualized treatment adjustments, and context-dependent clinical observations, making systematic decision support difficult. This study proposes an artificial intelligence-based framework for decision support and personalized intervention in TCM external therapy. The framework integrates patient characteristics, clinical symptoms, TCM diagnostic information, treatment modalities, intervention intensity, temporal changes, and outcome feedback into a structured representation, while considering heterogeneity and incompleteness in real-world clinical data. Machine learning methods are proposed to support patient stratification, intervention matching, and response prediction, with interpretable modeling and iterative clinical validation incorporated to reduce opaque recommendations. Rather than treating AI as a substitute for clinical judgment, the framework positions computational prediction as an auxiliary mechanism through which heterogeneous clinical experience and longitudinal patient information may be organized and evaluated. Challenges related to data standardization, treatment confounding, model transferability, interpretability, and ethical governance are also examined. Further research is needed to determine whether such systems can provide stable, clinically meaningful support across different TCM external-therapy settings.