Traditional Chinese medicine diagnostic prediction model for holistic syndrome differentiation based on deep learning

다국어 초록 (Multilingual Abstract)

Background: With the development of traditional Chinese medicine (TCM) syndrome knowledge accumulation and artificial intelligence (AI), this study proposes a holistic TCM syndrome differentiation model for the classification prediction of multiple TCM syndromes based on deep learning and accelerates the construction of modern foundational TCM equipment.


Methods: We searched publicly available TCM guidelines and textbooks for expert knowledge and validated these sources using ten-fold cross-validation. Based on the BERT and CNN models, with the classification constraints from TCM holistic syndrome differentiation, the TCM-BERT-CNN model was constructed, which completes the end-to-end TCM holistic syndrome text classification task through symptom input and syndrome output. We assessed the performance of the model using precision, recall, and F1 scores as evaluation metrics.


Results: The TCM-BERT-CNN model had a higher precision (0.926), recall (0.9238), and F1 score (0.9247) than the BERT, TextCNN, LSTM RNN, and LSTM ATTENTION models and achieved superior results in model performance and predictive classification of most TCM syndromes. Symptom feature visualization demonstrated that the TCM-BERT-CNN model can effectively identify the correlation and characteristics of symptoms in different syndromes with a strong correlation, which conforms to the diagnostic characteristics of TCM syndromes.


Conclusions: The TCM-BERT-CNN model proposed in this study is in accordance with the TCM diagnostic characteristics of holistic syndrome differentiation and can effectively complete diagnostic prediction tasks for various TCM syndromes. The results of this study provide new insights into the development of deep learning models for holistic syndrome differentiation in TCM.

다국어 초록 (Multilingual Abstract)

Background: With the development of traditional Chinese medicine (TCM) syndrome knowledge accumulation and artificial intelligence (AI), this study proposes a holistic TCM syndrome differentiation model for the classification prediction of multiple TCM syndromes based on deep learning and accelerates the construction of modern foundational TCM equipment.




Methods: We searched publicly available TCM guidelines and textbooks for expert knowledge and validated these sources using ten-fold cross-validation. Based on the BERT and CNN models, with the classification constraints from TCM holistic syndrome differentiation, the TCM-BERT-CNN model was constructed, which completes the end-to-end TCM holistic syndrome text classification task through symptom input and syndrome output. We assessed the performance of the model using precision, recall, and F1 scores as evaluation metrics.




Results: The TCM-BERT-CNN model had a higher precision (0.926), recall (0.9238), and F1 score (0.9247) than the BERT, TextCNN, LSTM RNN, and LSTM ATTENTION models and achieved superior results in model performance and predictive classification of most TCM syndromes. Symptom feature visualization demonstrated that the TCM-BERT-CNN model can effectively identify the correlation and characteristics of symptoms in different syndromes with a strong correlation, which conforms to the diagnostic characteristics of TCM syndromes.




Conclusions: The TCM-BERT-CNN model proposed in this study is in accordance with the TCM diagnostic characteristics of holistic syndrome differentiation and can effectively complete diagnostic prediction tasks for various TCM syndromes. The results of this study provide new insights into the development of deep learning models for holistic syndrome differentiation in TCM.

참고문헌 (Reference)

• 1 Gong L, "What does a TextCNN learn?"
• 2 Leung AYL, "Validation of evidence-based questionnaire for TCM syndrome differentiation of heart failure and evaluation of expert consensus" 18 (18): 70-, 2023
• 3 O’Brien KA, "Understanding the reliability of diagnostic variables in a Chinese Medicine examination" 15 (15): 727-734, 2009
• 4 Ma Y, "Traditional chinese medicine word representation model augmented with semantic and grammatical information" 13 (13): 296-, 2022
• 5 Wang Y, "The impact of artificial intelligence on traditional Chinese medicine" 49 (49): 1297-1314, 2021
• 6 Bai R, "The effectiveness and safety of Chinese Patent Medicines based on syndrome differentiation in patients following percutaneous coronary intervention due to acute coronary syndrome(CPM trial) : a nationwide Cohort Study" 109 : 154554-, 2023
• 7 Alshubaily I., "TextCNN with Attention for Text Classification"
• 8 Dong X, "TCMPR : TCM Prescription Recommendation Based on Subnetwork Term Mapping and Deep Learning" 2022 : 4845726-, 2022
• 9 Jiang M, "Syndrome differentiation in modern research of traditional Chinese medicine" 140 (140): 634-642, 2012
• 10 Liu R, "Study on syndrome differentiation strategy of phlegm and blood stasis syndromes of coronary heart disease based on expert consultation on medical cases" 10 (10): 9940-9952, 2021
• 11 Wang L, "Research progress in the treatment of slow transit constipation by traditional Chinese medicine" 290 : 115075-, 2022
• 12 Zhang CT, "Progress on prevention and treatment of cerebral small vascular disease using integrative medicine" 29 (29): 186-191, 2023
• 13 Cui Y, "Pre-training with whole word masking for Chinese BERT" 29 : 3504-3514, 2021
• 14 Jiang W, "Post-infectious cough of different syndromes treated by traditional Chinese medicines : a review" 14 (14): 494-510, 2022
• 15 Li M, "Personalized intelligent syndrome differentiation guided by TCM consultation philosophy" 2022 : 6553017-, 2022
• 16 Zhou S, "Perceptions of traditional Chinese medicine doctors about using wearable devices and traditional Chinese medicine diagnostic instruments : a mixed-methodology study" 8 : 20552076221102246-, 2022
• 17 Zhou L, "Natural language processing algorithms for normalizing expressions of synonymous symptoms in traditional Chinese medicine" 2021 : 6676607-, 2021
• 18 Hu C, "Multi-task joint learning model for Chinese word segmentation and syndrome differentiation in traditional Chinese medicine" 19 (19): 5601-, 2022
• 19 Ma S, "Machine learning in TCM with natural products and molecules : current status and future perspectives" 18 (18): 43-, 2023
• 20 Bao YF, "Intelligent acupuncture : data-driven revolution of traditional Chinese medicine" 3 (3): 271-284, 2023
• 21 Song C, "Diagnostic and categorization criteria for palpitations below the heart in traditional Chinese medicine : a delphi consensus study" 27 (27): 68-72, 2021
• 22 Liu J, "Development and validation of predictive model based on deep learning method for classification of dyslipidemia in Chinese medicine" 11 (11): 21-, 2023
• 23 Feng C, "Development and application of artificial intelligence in auxiliary TCM diagnosis" 2021 : 6656053-, 2021
• 24 Li D, "Deep learning and machine intelligence : new computational modeling techniques for discovery of the combination rules and pharmacodynamic characteristics of Traditional Chinese Medicine" 933 : 175260-, 2022
• 25 Zhang GD, "Comparison of syndrome differentiation and treatment system between Huangdi Neijing and Treatise on Cold Damage" 46 (46): 4699-4700, 2023
• 26 Gu TY, "Classifying Chinese medicine constitution using multimodal deep-learning model" 30 (30): 163-170, 2024
• 27 Zhang T, "Chinese herbal compound prescriptions combined with Chinese medicine powder based on traditional Chinese medicine syndrome differentiation for treatment of chronic atrophic gastritis with erosion : a multi-center, randomized, positive-controlled clinical trial" 17 (17): 142-, 2022
• 28 Matos LC, "Can traditional chinese medicine diagnosis be parameterized and standardized? A narrative review" 9 (9): 177-, 2021
• 29 Devlin J, "BERT: pre-training of deep bidirectional transformers for language understanding"
• 30 Li J, "Automatic classification framework of tongue feature based on convolutional neural networks" 13 (13): 501-, 2022
• 31 Vaswani A, "Attention is All you Need"
• 32 Zhang H, "Artificial intelligence-based traditional chinese medicine assistive diagnostic system : validation study" 8 (8): e17608-, 2020
• 33 Liu ZY, "Application of traditional Chinese medicine theory in modern traditional Chinese medicine nano-preparation : taking tumor treatment as an example" 48 (48): 1455-1462, 2023
• 34 Ding L, "Application of an extreme learning machine network with particle swarm optimization in syndrome classification of primary liver cancer" 19 (19): 395-407, 2021
• 35 Duan YY, "Application and development of intelligent medicine in traditional Chinese medicine" 41 (41): 1116-1122, 2021
• 36 김태훈 ; 강중원 ; Lee Myeong Soo, "AI Chat bot-ChatGPT-4 : a new opportunity and challenges in complementary and alternative medicine(CAM)" 12 (12): 100977-, 2023
• 37 Huang Z, "A traditional Chinese medicine syndrome classification model based on cross-feature generation by convolution neural network : model development and validation" 10 (10): e29290-, 2022
• 38 Liu Q, "A survey of artificial intelligence in tongue image for disease diagnosis and syndrome differentiation" 9 : 20552076231191044-, 2023
• 39 Zhang Y, "A sensitivity analysis of (and practitioners’ guide to)convolutional neural networks for sentence classification"