Advancements in Artificial Intelligence-Driven Diagnostic Models for Traditional Chinese Medicine

Traditional Chinese medicine (TCM) is an ancient medical system with distinctive ethnic characteristics. TCM diagnosis, underpinned by unique theoretical frameworks and methodologies, continues to play a significant role in contemporary healthcare. The four fundamental diagnostic methods, inspection, auscultation-olfaction, inquiry and palpation, are inherently subjective, relying on practitioner experience. Despite its unique advantages and practical value, TCM must still take advantage of modern advancements to enhance its effectiveness and accessibility. With the rapid development of computer technology, intelligent TCM diagnosis has emerged as a promising frontier. Integrating artificial intelligence (AI), particularly through large language models (LLMs), offers new avenues for enhancing TCM diagnostic practices. However, the systematic review and analysis of these technologies remains limited. This paper provides a comprehensive overview of the development and recent advancements in TCM diagnostic technologies, focusing on the applications of ML across various data modalities, and including images, text, and waveforms. Additionally, it explores the latest applications of LLMs within the TCM diagnostic field. Furthermore, the review discusses the prospects and challenges associated with AI-based TCM diagnosis. By systematically summarizing the latest research achievements and technological advancements, this study aims to provide directional guidance and decision support for future research and practical applications in the intersection of AI and TCM. Ultimately, this review seeks to foster the continued development and integration of intelligent TCM diagnosis into modern healthcare.