Revolutionizing traditional Chinese medicine research: Advancements in traditional machine learning and deep learning for AI-driven applications

Due to the numerous limitations of traditional methods in the research of traditional Chinese medicine (TCM), since entering the information age, artificial intelligence (AI) has not only played an increasingly important role in people's lives, but its components—traditional machine learning (ML) and deep learning (DL)—have also developed rapidly and demonstrated significant advantages and potential in TCM research. Therefore, they have been widely applied in areas such as TCM toxicity prediction, TCM efficacy assessment, targeted TCM development, identification and classification of TCM decoction pieces, authenticity and quality verification, quality control, and exploration of prescription composition rules. This article reviews and summarizes the specific applications of AI based on traditional ML and DL in the abovementioned research aspects in recent years, and identifies the challenges faced in the research. The aims are to summarize the current research status of AI in the field of TCM, clarify the technical methods employed, gain insights into the technical challenges encountered during the application process and ponder over them, and provide references and bases for future research in related aspects. At the same time, it explores the future applications trends of AI in TCM research, expounds on the importance of interdisciplinary cooperation in promoting the development of this field, looks ahead to its broad application prospects, and contributes to achieving more innovations and breakthroughs in TCM research. Application of artificial intelligence in TCM prediction, medicinal material identification, quality control, and formula compatibility optimization. • Comprehensive review of AI across the entire TCM chain, from herb to prescription. • AI assistance has significantly reduced research time and costs. • Addressing TCM standardization and personalization via data-driven AI approaches. • AI assistance has overcome the limitations of traditional methods in TCM research. • Emphasizing interdisciplinary collaboration is key to AI-driven TCM innovation.