AI-Empowered Traditional Chinese Medicine (TCM): Opportunities and Challenges in the Era of Large Language Models

AI-Empowered Traditional Chinese Medicine (TCM): Opportunities and Challenges in the Era of Large Language ModelsWith the advent of the AI-empowered era, large language models (LLMs) are rapidly permeating various industries, and the field of Traditional Chinese Medicine (TCM) is no exception. According to incomplete statistics, more than 20 TCM-related large models have emerged since 2023

AI-Empowered Traditional Chinese Medicine (TCM): Opportunities and Challenges in the Era of Large Language Models

With the advent of the AI-empowered era, large language models (LLMs) are rapidly permeating various industries, and the field of Traditional Chinese Medicine (TCM) is no exception. According to incomplete statistics, more than 20 TCM-related large models have emerged since 2023. This phenomenon reflects the urgent need for TCM to leverage modern methods for its inheritance and innovation. While TCM large models have shown some potential in areas like image diagnosis and intelligent consultation, their digital exploration remains in its infancy due to the complexities of TCM theory and practice. Several experts point out that in drug development, the accumulation of basic clinical data and talent cultivation remain significant challenges. Greater participation from clinical research institutions and frontline doctors, coupled with in-depth collaboration with LLM manufacturers, and the full utilization of real-world clinical data, are crucial to realizing the full value of large models.

Application Exploration of AI-Intelligent TCM Diagnosis and Treatment

Driven by large models, AI technology is starting to penetrate TCM clinical practice. Many TCM hospitals have introduced AI-intelligent TCM diagnostic instruments that collect patient facial features and tongue images for AI-intelligent recognition. Combined with knowledge from databases, these instruments generate body constitution analysis reports and personalized health recommendations. Some commercial entities are also actively exploring AI applications in TCM. For example, a tea shop in Nanjing has introduced "Cyber TCM Doctor," utilizing AI tongue and pulse diagnosis technologies to recommend matching beverages to customers.

Professor Ge Guangbo, Executive Vice President of the Cross-Science Research Institute at Shanghai University of Traditional Chinese Medicine, stated, "Artificial intelligence technology provides us with more effective methodologies." The university is developing a modern TCM four-diagnosis instrument using AI, collecting clinical image information and using AI for syndrome differentiation. Future applications will guide precise TCM treatment based on syndrome subtypes. Professor Ge also emphasized, "The secret of TCM lies in quantification."

The Booming Development of TCM Large Models

Since 2023, numerous TCM large models have emerged, driven by significant investment from universities, research institutes, and major technology companies such as Huawei and Baidu. In academia, Fudan University and Tongji University collaborated to develop the Zhongjing TCM large language model; East China Normal University, Shanghai University of Traditional Chinese Medicine, East China University of Science and Technology, Naval Medical University, Lingang Laboratory, and the National Key Laboratory of Modern TCM of Huarun Jiangzhong jointly developed the "Digital Qihuang" TCM large model; and the National Supercomputing Tianjin Center collaborated with the Modern TCM Haihe Laboratory, Tianjin University of Traditional Chinese Medicine, Tianjin University, and the Xin Chuang Haihe Laboratory to develop the "Tianhe Lingshu" large model. In the corporate sector, Huawei and Tiens jointly built the "Digital Bencao" large model; and Chengdu University of TCM, Baidu, and Taiji Group jointly released the "Bencao Wisdom Library," a large model covering the entire TCM industry chain.

Experts generally believe that China's rich collection of classical TCM texts and accumulated physician experience provide a solid foundation for "AI+TCM." Professor Ge Guangbo noted, "Artificial intelligence can also generate new applications in the exploration of the pharmacologically active substances and mechanisms of action of TCM, and play a role in modern pharmaceutical development, discovering new indications and developing innovative drugs."

Relatively Easy Implementation of Assisted Diagnosis Applications

The rise of TCM large models stems from the urgent need for modernization in TCM inheritance and innovation. The "Opinions on Promoting the Inheritance and Innovation Development of TCM" issued in 2019 clearly proposed using information technology to support the construction of the TCM service system, implementing the "Internet + TCM Health Services" action, and improving the informationization level of TCM. In August of this year, the National Administration of TCM and the National Data Administration jointly issued "Several Opinions on Promoting the Development of Digital TCM," proposing 20 measures to promote the development of digital TCM.

Academician Chen Kaixian of the Chinese Academy of Sciences, at a recent academic conference on network pharmacology and artificial intelligence, pointed out that TCM has accumulated a large amount of complex data, and modern science and technology can be fully applied to the inheritance and innovative development of TCM. Artificial intelligence will drive technological transformation in the innovative development of TCM.

Currently released large models are primarily used for assisted consultation, with some exploration in TCM originality and research and development. Experts believe that due to the development of wearable devices and the accumulation of imaging data, the application of artificial intelligence in assisted diagnosis is easier to break through and play a greater role. However, the industry is more hopeful that AI will play a greater role in TCM research and development. Academician Chen Kaixian introduced that Nanjing University of Information Science & Technology and Beijing Jiaotong University have conducted interdisciplinary collaborative research, using network pharmacology to reveal the common mechanisms of TCM syndrome differentiation and treatment, providing new insights into the scientific connotation of TCM syndrome differentiation and treatment, and exploring the application of artificial intelligence and big data in addressing problems in the TCM field.

Professor Ge Guangbo pointed out that compared with chemical drugs, TCM usually acts on multiple drug targets through multiple components and multiple signaling pathways, producing an integrated synergistic effect. This effect makes it difficult to screen out a single component to replace the pharmacological activity of the entire compound preparation in the research of new TCM drugs, thus requiring innovation in methodology and research strategies.

High-Quality Clinical Data: A Key Bottleneck in the Digitalization of TCM

Huawei Cloud and Tiens, together with experts and scholars from multiple universities and research institutes, launched the Digital TCM Innovation Alliance on November 24, 2023, aiming to promote the application and development of modern technology in the innovative field of TCM. Li Yin, General Manager of Huawei Cloud's Scientific Computing Domain, mentioned that Huawei Cloud hopes to integrate a large platform to attract global users and developers to participate and simplify related work. The "Digital Bencao" large model aims to promote the development of TCM research and development.

The TCM field has accumulated thousands of years of experience and literature, but its mechanisms of action on the human body are still not clear. Currently, the TCM large models proposed in China generally use traditional TCM experience knowledge graphs as training data. However, how to use modern scientific methods to analyze the mechanisms of action of TCM, achieve a one-to-one correspondence between TCM and target databases, and scientifically optimize and improve the clinical efficacy of TCM are key issues that need to be addressed.

Several experts said that in the digitalization process of the TCM industry, the underlying database and talent cultivation are the two biggest difficulties, requiring the joint efforts of the TCM industry, large model manufacturers, and all sectors of society. Professor Fan Xiaohui, director of the Yangtze River Delta Smart Green Oasis Innovation Center at Zhejiang University, pointed out, "In fact, it may be that our industry itself is not yet ready." He stressed that high-quality big data is the key. Currently, the TCM field is relatively rich in classical text data, but clinical trial data still lacks effective sorting and standardization. Data dispersion, lack of standardization, the difficulty of combining data with clinical practice, data personalization, patient privacy, and ethical factors all hinder the effective use of data.

Professor Cao Yongbing, director of the Shanghai Institute of Integrated Traditional Chinese and Western Medicine for Vascular Diseases, as one of the initiators of the "Digital TCM Innovation Alliance," emphasized the importance of close cooperation between large model manufacturers and clinical units and frontline TCM practitioners to promote the collection and organization of high-quality TCM clinical diagnostic data to provide effective support for TCM large models and enable them to play an important role in new drug development and TCM clinical diagnosis and treatment. Professor Cao mentioned that clinical data must be real and effective, and many large models lack the participation of clinical teams in the input of underlying data, disconnecting from real scientific research. He vividly compared it to, "Without data, having a pile of large models is like building a lot of unfinished buildings."

In addition to high-quality clinical data, talent is also a major bottleneck. Some universities have begun exploring interdisciplinary research between artificial intelligence and TCM. Shanghai University of Traditional Chinese Medicine is preparing China's first AI research institute in the field of TCM, and the pharmaceutical discipline of Zhejiang University also focuses on using interdisciplinary technologies to promote the construction of a new drug research system. Professor Fan Xiaohui said that to truly achieve the digitalization of TCM, it is necessary to cultivate compound talents who understand both artificial intelligence technology and TCM scientific issues, and currently, such talents are very scarce. He stressed that the talent cultivation system needs to be reformed to meet new needs, which requires the joint efforts of the TCM industry, the state, and all industries.

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])