Analytical Methods And Application Of TCM Clinical Drug Compatibility

Traditional Chinese medicine is a medical system with a unique theoretical system and original thinking in China.In the long-term clinical practice research,many innovative theories such as prescription compatibility and drug compatibility theory have been formed.Achieve reasonable drug compatibility based on prescription treatment system is of great significance to the discovery the rule of prescription compatibility and individualized treatment research.Among them,drug compatibility and dose compatibility are two important factors.Study the compatibility mode of two factors,so as to realize better drug compatibility in order to achieve personalized treatment,is the main goal of the compatibility of traditional Chinese medicine research.At present,with the accumulation of large-scale clinical data,how to conduct compatibility analysis of traditional Chinese medicine in a human-machine combination is the current research hotspot and the main research direction of TCM clinical prescription analysis.Combining the clinical data of TCM and the basic prescription database data,this article conducts research on compatibility of traditional Chinese medicine in three aspects,including the compatibility analysis and application of core drugs based on complex network,and drug compatibility based on dose analysis,and the analysis of drug roles based on sequence annotation model.The main research results include the following three aspects:(1)The analysis method and application of the core drugs compatibility of Chinese medicine:Using complex network rapid module partitioning method and hierarchical network analysis method to analyze the compatibility of the core drugs,and combine with the clinical data of an old Chinese medicine record of nearly 10,000 for demonstration analysis and application research.In the end,we get 126 prescription modules and 4 drug modules which are confirmed by the famous old Chinese medicine and his disciples.The combination of the compound core drugs excavated by this method embodies the general medicine characteristics of the prescription of the famous old Chinese medicine,and provides an effective method for mining and analyzing the experience of the famous old Chinese medicine.(2)Proposed Copula-based drug-pair correlation analysis method:dose is an important information for compatibility of traditional Chinese medicine,but dose information is rarely considered in the compatibility analysis of traditional Chinese medicine,which greatly affects the compatibility analysis results.This paper proposes a drug-pair compatibility analysis method that combines association rules and Copula function.We use the Apriori algorithm in the association rules to determine common drug pairs,and use Copula to construct a joint distribution function based on drug dose to determine the relevant structure of drug pairs,at last,we use Kendall and Spearman correlation coefficients to calculate correlation values between drug pairs.The experiment analysis of nearly 10,000 clinical prescription data shows that the combination of qualitative and quantitative methods can be used to analyze the compatibility of drugs,and it can better analyze drug compatibility so as to acquire more clinical value of drug knowledge.(3)Put forward the role analysis method of Chinese medicine based on sequence annotation model:The different roles of the drug in prescription such as Jun Chen Zuo Shi are important information for the action of the drug,but the information in the conventional clinical prescription is only in the memory of the doctor,and it is not recorded in the clinical data.On the basis of dose standardization,this paper proposes an automatic annotation method based on conditional random fields(CRF)for automatic identification of drug roles in a given compound.Based on the 258 prescription data of prescription textbook,the standardized dose and drug efficacy were used as feature to training model.The average accuracy rate of 10-fold cross-validation was 90%,and the results were verified with 20 clinical data of a famous doctor.The result shows that when the ratio of drug efficacy and the symptom correlation value and standardization dose was 8:2,the drug was ranked,the accuracy rate of the result was 65.87%,which provided a method and a train of thought for compound drug compatibility and development of new drugs.