Patient-Conditioned Dual Hypergraph Reasoning for Auditable Traditional Chinese Medicine Prescription Support

2026-07-04

AI Summary

This study addresses the challenge of generating individualized traditional Chinese medicine (TCM) prescriptions, which require syndrome differentiation tailored to each patient, yet current language models lack auditability and static knowledge bases struggle with dynamic adaptation. The authors propose the first patient-condition-driven dual hypergraph reasoning framework: the first hypergraph enables traceable inference from symptoms to syndromes and treatment principles, while the second integrates syndromes, therapeutic methods, disease context, and dosage priors to generate personalized prescriptions. Both hypergraphs are dynamically weighted according to patient-specific representations. Combining MacBERT-based semantic encoding with clinical evidence retrieval, the model achieves a syndrome recognition accuracy of 0.8297 on TCM-SD and a herb-level F1 score of 0.3111 on TCM-BEST4SDT, approaching the theoretical performance upper bound. Practical utility is further validated through audits of 50 real clinical cases.

Abstract

Traditional Chinese medicine (TCM) prescription support requires patient-specific reasoning from clinical narratives to syndromes, treatment principles, herbs, and doses. Direct language-model generation can produce fluent prescriptions, but its decisions are difficult to audit against explicit clinical evidence. Static TCM knowledge resources provide useful priors, but they cannot determine which diagnostic and prescription relations should be emphasized for an individual patient. We propose a patient-conditioned dual hypergraph framework for auditable TCM prescription support. The first hypergraph organizes symptom, tongue, pulse, and other clinical evidence around syndrome and treatment-principle reasoning. The second hypergraph organizes syndrome, treatment, disease-context, herb, retrieval, and dose-prior evidence for prescription construction. Unlike static knowledge graphs or fixed hypergraphs, both hypergraphs are dynamically weighted by the patient representation. This design enables individualized activation of diagnostic and prescription paths, supporting personalized syndrome differentiation and herb-dose recommendation while preserving case-level auditability. Experiments on TCM-SD show that dynamic weighting in the first hypergraph improves MacBERT syndrome differentiation to 0.8297 accuracy and 0.3288 macro-F1. On TCM-BEST4SDT, the second hypergraph achieves the best mean Herb-F1 of 0.3111 across three seeds, and the full connected pipeline reaches 0.3074 Herb-F1, close to the oracle setting. A 50-case real-world CAP audit further suggests practical review potential, while highlighting the need for prospective dose-safety validation.

Problem

• auditable prescription
• patient-specific reasoning
• TCM syndrome differentiation
• herb-dose recommendation
• clinical evidence

Innovation

• patient-conditioned hypergraph
• auditable reasoning
• TCM prescription support
• dynamic knowledge graph
• syndrome differentiation