Related Experiment Video
Updated: Aug 6, 2026
08:43
A Study on an Intelligent Diagnosis and Treatment Assistant System for Acupuncture in Diminished Ovarian Reserve Based on a Knowledge Graph
Published on: May 29, 2026
Modeling symptom-acupoint interactions via a heterogeneous graph learning framework for intelligent acupoint
Wei Lin1,2,3,4, Jiaqi Chen1, Zhuo Chen4
1School of Acupuncture, Moxibustion and Tuina, Chengdu University of Traditional Chinese Medicine, Chengdu, 610075, China.
Chinese Medicine
|July 17, 2026
Summary
This study introduces a novel Graph Neural Network-BERT-Attention (GNN-BERT-Attention) model for accurate acupuncture point prediction. The framework effectively models complex symptom-acupoint relationships, improving treatment precision.
Area of Science:
- Computational Medicine
- Traditional Chinese Medicine
- Artificial Intelligence in Healthcare
Background:
- Acupuncture prescriptions rely on complex interactions between symptoms and acupoints, honed over centuries.
- Current computational methods for acupoint recommendation face challenges due to sparse data and inadequate modeling of symptom-acupoint relationships.
Purpose of the Study:
- To develop an advanced computational framework for predicting compatible acupoints.
- To address data sparsity and improve the accuracy of acupoint recommendations in traditional Chinese medicine.
Main Methods:
- Introduced a Graph Neural Network-BERT-Attention (GNN-BERT-Attention) framework.
- Utilized heterogeneous feature interaction learning for modeling symptom-acupoint relationships.
- Employed neural collaborative filtering with label-aware fusion and Focal Loss for robust predictions.
Main Results:
- The GNN-BERT-Attention model significantly outperformed State-of-the-Art baselines in precision, recall, and robustness.
- Ablation studies confirmed the effectiveness of individual architectural components.
- A web-based system demonstrated real-time, interpretable acupoint recommendations, validating clinical applicability.
Conclusions:
- The study enhances acupoint prediction performance, improving acupuncture treatment efficiency and precision.
- Provides a theoretical foundation for optimizing traditional Chinese medicine prescriptions.
- Advances evidence-based practices in traditional Chinese medicine interventions.
