Large AI models improve traditional Chinese medicine prescription decisions
Lingdan-V2: Large language models for clinically grounded reasoning in traditional Chinese medicine
Computation and LanguageArtificial Intelligence
Summary
It can be hard for computer programs to help with traditional Chinese medicine because it involves many complex steps like diagnosing and choosing herbal medicines with correct dosages. The authors created a set of large language models specifically trained on Chinese medicine texts and clinical records. Their approach uses a special training process that helps the models learn how to reason through prescriptions and improve recommendations. Tests show the largest model can produce herbal prescriptions more accurately than earlier systems, supporting doctors in making better treatment choices.
What this means in practice
- •For traditional chinese medicine practitioners: Assist practitioners in generating accurate herbal prescriptions tailored to patient syndromes and dosages based on clinical records and TCM knowledge.
- •For healthcare ai developers: Build clinical decision support tools that incorporate structured reinforcement learning for better prescription recommendation in complementary and integrative medicine.$Commercial implications: Enables creation of AI products that provide reliable traditional medicine prescription support, offering new clinical software solutions.
Authors
Rui Hua, Yu Wei, Ziri Jin, Zhuo Sun, Kai Chang, Jianan Xia, Zixin Shu, Xiaodong Li, Dongmei Jia, Fei Dong, Runshun Zhang, Jian Yu, Hao Xu, Haibin Yu, Jiansheng Li, Baoyan Liu, Wenjia Wang, Xuezhong Zhou
Abstract
Large language models have advanced medical reasoning, but clinically grounded reasoning in traditional Chinese medicine (TCM) remains challenging because prescription decisions require integrating domain knowledge, syndrome differentiation, treatment principles, herbal prescription composition, and dosage determination. We develop a TCM language model family comprising models with 4 billion, 8 billion, and 14 billion parameters. The artificial intelligence contribution is a staged clinical alignment framework integrating large scale continued pretraining on TCM and clinical records, complementary reasoning supervision, and prescription-centered reinforcement learning (RL). We further formulate herbal prescription generation as a structured RL problem with prescription-level feedback, enabling direct optimization of clinical decision outputs beyond knowledge reproduction and generic instruction following. The engineering application is TCM clinical decision support, particularly herbal prescription recommendation. We evaluate the models on a public benchmark containing 25,620 expert-verified items across 13 TCM knowledge and clinical reasoning tasks and on a private retrospective dataset of 4,348 spleen and stomach disease prescription cases. The 14 billion parameter reinforcement aligned model achieves a prescription generation F1 of 37.0 on the public benchmark, compared with 32.3 for DeepSeek-R1 with 671 billion total parameters, and a herb-level F1 of 52.8 on the clinical dataset with improved dosage agreement. Reinforcement post-training consistently improves prescription recommendation across all scales, while physician-source analysis shows that expert heterogeneity affects reward stability. These findings demonstrate a scalable pathway from domain grounding and reasoning supervision to clinically oriented prescription decision support.