Intelligent Prescription-Generating Models of Traditional Chinese Medicine Based on Deep Learning

Abstract

Objective: This study aimed to construct an intelligent prescription-generating (IPG) model based on deep-learning natural language processing (NLP) technology for multiple prescriptions in Chinese medicine. Materials and Methods: We selected the Treatise on Febrile Diseases and the Synopsis of Golden Chamber as basic datasets with EDA data augmentation, and the Yellow Emperor's Canon of Internal Medicine, the Classic of the Miraculous Pivot, and the Classic on Medical Problems as supplementary datasets for fine-tuning. We selected the word-embedding model based on the Imperial Collection of Four, the bidirectional encoder representations from transformers (BERT) model based on the Chinese Wikipedia, and the robustly optimized BERT approach (RoBERTa) model based on the Chinese Wikipedia and a general database. In addition, the BERT model was fine-tuned using the supplementary datasets to generate a Traditional Chinese Medicine-BERT model. Multiple IPG models were constructed based on the pretraining strategy and experiments were performed. Metrics of precision, recall, and F1-score were used to assess the model performance. Based on the trained models, we extracted and visualized the semantic features of some typical texts from treatise on febrile diseases and investigated the patterns. Results: Among all the trained models, the RoBERTa-large model performed the best, with a test set precision of 92.22%, recall of 86.71%, and F1-score of 89.38% and 10-fold cross-validation precision of 94.5% ± 2.5%, recall of 90.47% ± 4.1%, and F1-score of 92.38% ± 2.8%. The semantic feature extraction results based on this model showed that the model was intelligently stratified based on different meanings such that the within-layer's patterns showed the associations of symptom–symptoms, disease–symptoms, and symptom–punctuations, while the between-layer's patterns showed a progressive or dynamic symptom and disease transformation. Conclusions: Deep-learning-based NLP technology significantly improves the performance of IPG model. In addition, NLP-based semantic feature extraction may be vital to further investigate the ancient Chinese medicine texts.

The combination of molecular docking and network pharmacology reveals the molecular mechanism of Danggui Niantong decoction in treating gout

“…it is a classic prescription for the treatment of gout. [13] In this study, we found that the core active components of DGNTD in the treatment of gout include quercetin, kaempferol, formononetin, luteolin and wogonin. Quercetin is the active compound of Licorice (Gancao), Herba Artemisiae (Yinchen) and Sophora flavescens (Kushen), kaempferol is the active compound of Licorice (Gancao), Ginseng (Renshen) and Sophora flavescens (Kushen), formononetin is the active compound of Licorice (Gancao), Radix puerariae (Gegen) and Sophora flavescens (Kushen), luteolin is the active compound of Radix Atractylodis (Cangzhu), and baicalein is the active compound of Saposhnikoviae Radix (Fangfeng), Radix Atractylodis (Cangzhu) and Radix Scutellariae (Huangqin).…”

“…DGNTD has been widely used in clinical treatment of gout in China and achieved satisfactory results. [12,13] Previous studies have shown that DGNTD can improve synovitis in arthritic rats by inhibiting synovial hyperplasia, infiltration of inflammatory cells, and destruction of cartilage and bone during inflammation. [14] Recently, DGNTD has been proved to effectively alleviate the degree of inflammatory swelling of paws in AGA mice, down-regulate the expression of inflammatory factors IL-1β, TNF-α and iNOSmRNA, and reduce the concentration of serum uric acid.…”

Effect of nursing intervention based on continuity model on psychological stress state of patients with rheumatoid arthritis

“…During the nursing intervention based on the continuity model, the self-developed multifunctional arm frame is used to properly elevate the affected limb and assist joint function rehabilitation training, so as to improve the functional rehabilitation effect of the affected upper limb and ultimately improve the quality of life. [21] Nursing intervention based on continuity model attaches importance to music therapeutic touch and music rehabilitation exercise, so that patients can carry out rehabilitation training accompanied by music, which does not rely on any drugs, and use the special relationship between music and people to improve the body's health status. As an ideal "natural therapy," musical therapeutic touch and musical rehabilitation exercise directly affect brain stem reticular structure, limbic system, and hypothalamus through the special physiological effect of music, stimulate pituitary secretion and release of endorphins, so as to effectively improve the rehabilitation effect of upper limb function and improve the quality of life.…”

Roadmap for modern development of Chinese medicine: Traditional, evidence-based, and digital-intelligent paths

“…It would ensure the quality of medications and meet clinical needs by enhancing the intelligence and informatization of instruments on the existing technological foundation, achieving intelligent pharmaceutical manufacturing. [43] 3. Looking toward the future…”