ABSTRACT
Chinese medicine (CM) diagnosis intellectualization is one of the hotspots in the research of CM modernization. The traditional CM intelligent diagnosis models transform the CM diagnosis issues into classification issues, however, it is difficult to solve the problems such as excessive or similar categories. With the development of natural language processing techniques, text generation technique has become increasingly mature. In this study, we aimed to establish the CM diagnosis generation model by transforming the CM diagnosis issues into text generation issues. The semantic context characteristic learning capacity was enhanced referring to Bidirectional Long Short-Term Memory (BILSTM) with Transformer as the backbone network. Meanwhile, the CM diagnosis generation model Knowledge Graph Enhanced Transformer (KGET) was established by introducing the knowledge in medical field to enhance the inferential capability. The KGET model was established based on 566 CM case texts, and was compared with the classic text generation models including Long Short-Term Memory sequence-to-sequence (LSTM-seq2seq), Bidirectional and Auto-Regression Transformer (BART), and Chinese Pre-trained Unbalanced Transformer (CPT), so as to analyze the model manifestations. Finally, the ablation experiments were performed to explore the influence of the optimized part on the KGET model. The results of Bilingual Evaluation Understudy (BLEU), Recall-Oriented Understudy for Gisting Evaluation 1 (ROUGE1), ROUGE2 and Edit distance of KGET model were 45.85, 73.93, 54.59 and 7.12, respectively in this study. Compared with LSTM-seq2seq, BART and CPT models, the KGET model was higher in BLEU, ROUGE1 and ROUGE2 by 6.00-17.09, 1.65-9.39 and 0.51-17.62, respectively, and lower in Edit distance by 0.47-3.21. The ablation experiment results revealed that introduction of BILSTM model and prior knowledge could significantly increase the model performance. Additionally, the manual assessment indicated that the CM diagnosis results of the KGET model used in this study were highly consistent with the practical diagnosis results. In conclusion, text generation technology can be effectively applied to CM diagnostic modeling. It can effectively avoid the problem of poor diagnostic performance caused by excessive and similar categories in traditional CM diagnostic classification models. CM diagnostic text generation technology has broad application prospects in the future.
Subject(s)
Humans , Medicine, Chinese Traditional , Pattern Recognition, Automated , Asian People , Language , LearningABSTRACT
Objective@#To provide ideas for the modernization of pulse diagnosis in traditional Chinese medicine (TCM) by comparing and analyzing the current status and trends of modern research on pulse diagnosis in China and abroad, using bibliometric and visualization software.@*Methods@#Modern research literature on pulse diagnosis was searched in China National Knowledge Infrastructure (CNKI) database from the foundation to May 31, 2023, and in Science Citation Index Expanded (SCIE) from January 1, 2003, to May 31, 2023. After further screening, Microsoft Excel 2019 was used for statistical analysis of publication volume, and CiteSpace (6.1.R6) and VOSviewer (1.6.20) softwares were employed for visual analysis of journals, countries/regions, authors, institutions, keywords, etc.@*Results@#This study included a total of 764 articles in Chinese and 1 459 articles in English. The publication trend of pulse diagnosis research in SCIE database showed an overall fluctuating upward trend, while it exhibited a fluctuating downward trend after 2007 in CNKI database. The volume of English research literature has consistently exceeded that of Chinese literature since 2009. Publications on pulse diagnosis research involved 74 countries/regions. The related journals covered various disciplinary fields, including mathematics, physics, chemistry, and computer science. The most prolific author in CNKI database was WANG Yiqin (Shanghai University of Traditional Chinese Medicine), while the highest number of publications was attributed to ZHANG David (The Hong Kong Polytechnic University) in SCIE database. High-volume institutions in pulse diagnosis research in China and abroad were predominantly TCM research institutions. However, comprehensive universities and other research institutions also made noteworthy contributions. In recent years, hot topics in the modernization of pulse diagnosis research in China included pulse waves, sensors, and artificial intelligence. Foreign research focused on pulse diagnosis systems, sensors, pulse feature extraction, pulse signal analysis, pulse detection, and efficiency of use.@*Conclusion@#Chinese scholars have shown notable participation and emphasis in the modernization research of pulse diagnosis, involving a wide range of disciplinary fields and indicating a characteristic of multidisciplinary cross-fusion development. The hotspots andtrends in the modernization of pulse diagnosis research primarily concentrate on the study of pulse condition and signal acquisition, the integration, development, and optimization of various algorithms with pulse diagnosis equipment, and the practical application research of existing objectified outcomes of pulse diagnosis.
ABSTRACT
Colorectal cancer is one of the most common malignant tumors of digestive tract. In 2020， 1.93 million new cases of colorectal cancer were diagnosed globally， ranking third in the global incidence spectrum， and 930 000 new deaths were reported， ranking second in the global cause of death spectrum. Meanwhile， the medical cost of metastatic colorectal cancer is the highest among all stages. A large number of studies have demonstrated that traditional Chinese medicine（TCM） treatment can bring clinical benefits to patients with metastatic colorectal cancer with unique efficacy. In order to further standardize the TCM diagnosis and treatment for metastatic colorectal cancer and improve the level of TCM diagnosis and treatment， Xiyuan Hospital， China Academy of Chinese Medical Sciences， together with other relevant units in China， according to the guideline development process of the World Health Organization Handbook for Guideline Development and the relevant requirements of the Clinical Evidence Grading Criteria on TCM Based on Evidence Body， the Regulations for Group Standards of China Association of Chinese Medicine and others， combined with the characteristics of TCM diagnosis and treatment and the actual situation in China， the Guidelines for TCM Diagnosis and Treatment of Metastatic Colorectal Cancer was developed in accordance with the Catalogue of TCM Diagnosis and Treatment Plans for 105 Diseases in 24 Specialties issued by Department of Medical Administration of National Administration of TCM.
ABSTRACT
The purpose of this article is to evaluate the clinical effect of clinical practice guideline on traditional Chinese medicine therapy alone or combined with antibiotics for acute tonsillitis. The applicability and application of the Guideline were evaluated based on the clinicians by using the electronic questionnaire. Questionnaires about 538 on application evaluation and 502 questionnaires on applicability evaluation were completed from April 28 to July 9, 2018. The subjects in the questionnaires include the clinicians with junior title, intermediate title, and senior title that have used this Guideline. The descriptive statistical analysis of the collected questionnaire was carried out. In the applicability evaluation, according to the classification of professional titles, the application rate was highest in intermediate title clinicians (26.77%), followed by junior (23.98%) and deputy senior (19.33%) professional title clinicians. In the quality evaluation, the rationality of application scope (98.61%) and the terminology accuracy (98.81%) scores were higher, and the rationality of differentiation and classification (96.05%) was the lowest. The applicability evaluation suggested that clinicians believed this Guideline had high safety (98.42%), reasonable content (98.03%), significant effect (99.6%), reduced use of antibiotics in Western medicine (93.89%), and a high applicability ratio (96.44%). In the application evaluation, Department of lung disease showed the highest application rate (44.24%); rationality of the Guideline was more than 97% in treatment rules and prophylaxis except the syndrome differentiation (92.75%); a high ratio of clinicians believed the recommended scheme was good: curative effect 97.4%, safety 97.59%, and economy 93.87%. The study shows that the clinical practice guideline on traditional Chinese medicine therapy alone or combined with antibiotics for acute tonsillitis is of good quality, high clinical use and good effect. It can be used as a standardized treatment scheme for acute tonsillitis in traditional Chinese medicine. But there are some unsuitable contents and need to be further improved. The Guideline should strengthen the revision on differentiation of symptoms and signs as well as prophylaxis.
Subject(s)
Humans , Anti-Bacterial Agents , Medicine, Chinese Traditional , Practice Guidelines as Topic , Surveys and Questionnaires , Syndrome , Tonsillitis , Drug TherapyABSTRACT
Through several thousand years’ inheritance and development by Chinese people, traditional Chinese medicine (TCM) has formed its unique theoretic system, whose efficacy has been widely accepted. However, because TCM theory focuses on the relationships among syndromes, treatment and efficacy, instead of the cause-and-effect relationship explored by modern natural science, the scientificity of TCM has always been questioned. In recent years, because virtual-world clinical research mode and the concept of “big data” were emphasized, increasing researchers began to put their research emphasis on the correlativity between intervening measures of diseases and outcome indicators. This change and the advancement of computer data mining and analysis technology, bring great opportunities for the further development of TCM theory and practice. This article concluded data mining technology used in TCM diagnosis and treatment in recent years, such as clustering analysis, decision tree, Bayesian network, neural network and multi-instance learning, which showed how to apply these methods to reveal rules of TCM diagnosis and treatment from a large number of TCM syndrome data, find knowledge hidden in data, and show TCM effectiveness supported by data.