“…At the drug/formula-level, NLP techniques are mainly used to mine the relationships between different clinical symptoms related to diseases/syndromes from prior knowledges. According to the TCM clinical diagnosis and treatment theory, we could not only infer the similarities of clinical symptoms by mining clinical phenotype entries [ 18 ], but also infer the relationships between syndromes and phenotypes [ 19–23 ]. At the level of drugs/formulas, the structural similarity relationships between the components contained in formulas are mainly mined from databases such as PubChem [ 24 ], ChEMBL [ 25 ], CDCDB [ 26 ] and DrugCombDB [ 27 ], as well as the relationships between the herbs contained in formulas and their properties (e.g.…”
Network pharmacology (NP) provides a new methodological perspective for understanding traditional medicine from a holistic perspective, giving rise to frontiers such as traditional Chinese medicine network pharmacology (TCM-NP). With the development of artificial intelligence (AI) technology, it is key for NP to develop network-based AI methods to reveal the treatment mechanism of complex diseases from massive omics data. In this review, focusing on the TCM-NP, we summarize involved AI methods into three categories: network relationship mining, network target positioning and network target navigating, and present the typical application of TCM-NP in uncovering biological basis and clinical value of Cold/Hot syndromes. Collectively, our review provides researchers with an innovative overview of the methodological progress of NP and its application in TCM from the AI perspective.
“…At the drug/formula-level, NLP techniques are mainly used to mine the relationships between different clinical symptoms related to diseases/syndromes from prior knowledges. According to the TCM clinical diagnosis and treatment theory, we could not only infer the similarities of clinical symptoms by mining clinical phenotype entries [ 18 ], but also infer the relationships between syndromes and phenotypes [ 19–23 ]. At the level of drugs/formulas, the structural similarity relationships between the components contained in formulas are mainly mined from databases such as PubChem [ 24 ], ChEMBL [ 25 ], CDCDB [ 26 ] and DrugCombDB [ 27 ], as well as the relationships between the herbs contained in formulas and their properties (e.g.…”
Network pharmacology (NP) provides a new methodological perspective for understanding traditional medicine from a holistic perspective, giving rise to frontiers such as traditional Chinese medicine network pharmacology (TCM-NP). With the development of artificial intelligence (AI) technology, it is key for NP to develop network-based AI methods to reveal the treatment mechanism of complex diseases from massive omics data. In this review, focusing on the TCM-NP, we summarize involved AI methods into three categories: network relationship mining, network target positioning and network target navigating, and present the typical application of TCM-NP in uncovering biological basis and clinical value of Cold/Hot syndromes. Collectively, our review provides researchers with an innovative overview of the methodological progress of NP and its application in TCM from the AI perspective.
“…The results showed that both CNN and FastText could achieve good results. Mucheng et al [30] proposed a large data set for differentiation of TCM syndrome called TCM-SD. In addition, they pre-trained the ZY-BERT model on TCM-SD.…”
Section: Related Workmentioning
confidence: 99%
“…Recent SD studies. ZY-BERT [30]: Based on TCM-SD pre-training language model, the classification results are obtained directly with CLS symbols. SDTM [31]: Method of TCM syndrome differentiation based on topic model.…”
Section: Deep Learningmentioning
confidence: 99%
“…At the same time, there are many rare domain words in TCM text, and it is difficult to obtain accurate semantic vector representation by the existing language model. In addition, TCM syndrome differentiation results are diverse, with 148 standard syndrome types in TCM-SD [4]. Therefore, the task of differentiation of TCM syndrome based on diagnostic and treatment texts is challenging.…”
Syndrome differentiation (SD) is a basic task in TCM (Traditional Chinese Medicine) diagnosis and treatment. TCM syndrome differentiation is very complex and time-consuming. Meanwhile, the accuracy of the results depends on the experience of TCM practitioners. To help TCM practitioners differentiate syndrome more quickly, we propose a syndrome differentiation method of deep learning based on multi-feature fusion. We extracted char, word and POS (Part of Speech) from TCM diagnosis and treatment records. The vector representation of char feature is obtained by ZY-BERT (Zhong Yi BERT), ZY-BERT was pre-trained on large datasets of TCM-SD (TCM Syndrome Differentiation). The vector representation of word and POS is obtained by Word2vec (Word to vector). We construct text graphs of char, word and POS according to context. GCN (Graph Convolutional Networks) is used to extract spatial structure information between multiple features to achieve multi-feature fusion. The experiment was carried out on TCM-SD. The experimental results showed that the accuracy of the proposed method was 81.52%, which was better than the comparison method. This method is helpful in the development of TCM modernization.
BACKGROUND
Traditional Chinese Medicine (TCM), as an ancient medical system distinct from modern medicine, plays a crucial role in maintaining people's health.The accuracy of Traditional Chinese Medicine (TCM) treatment with syndrome differentiation outcomes is closely related to the experience of physicians. However, primary healthcare practitioners often lack experience, and the relatively low accuracy of intelligent TCM syndrome differentiation results is one of the current challenges. Therefore, a key research focus in the field of intelligent TCM lies in how to quickly and accurately differentiate patients while simultaneously enhancing patient satisfaction.
OBJECTIVE
The classification of traditional Chinese medicine (TCM) syndrome is an integral component of the TCM diagnostic system. Leveraging artificial intelligence (AI) technology to explore TCM syndrome classification models and enhance their performance holds the potential to extend into various other applications within the field of TCM.
METHODS
This study employs a residual structural graph convolutional neural network to capture deep features among data nodes, while integrating syndrome-related knowledge graphs to assist in consolidating syndrome embedding representations. This enhances the correlation between symptoms and syndrome by proposing the augmentation of status element weights as their bridge, integrating multi-layer information representations. Furthermore, a multilayer perceptron is utilized for syndrome classification.
RESULTS
The experimental results demonstrate that the proposed KGRGCN model achieves a precision of 75.43%, an accuracy of 74.93%, a recall of 76.91%, and an F1-score of 75.91%.
CONCLUSIONS
The proposed syndrome classification method outperforms several popular classification methods, including Support Vector Machine, TextCNN, and Random Forest.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.