Research On Key Technologies For Knowledge Mining Of Clinical Use Of Traditional Chinese Medicine In Yiyang Central Hospital

Objective: under the guidance of the theory of traditional Chinese medicine and the thought of evidence-based medicine,taking the mining of clinical big data of traditional Chinese medicine in Yiyang Central Hospital as the starting point,to explore the method and technology of mining clinical big data of traditional Chinese medicine,so as to make the diagnosis and treatment knowledge deposited in the data explicit and lay the foundation for better supporting doctors to carry out clinical practice of traditional Chinese medicine.Methods: collect the clinical data of patients with traditional Chinese medicine in Yiyang Central Hospital from January 2014 to December 2018,such as gender,age,nationality,hospitalization history,complications,disease diagnosis,hospitalization days,test items,inspection items,drug allergy history,TCM diagnosis syndrome type,proprietary Chinese medicine,prescription,single medicine,etc.,build a knowledge database of traditional Chinese medicine for Yiyang Central Hospital,and set up a knowledge database of traditional Chinese medicine in Ju On the platform of pter notebook,we use Python artificial intelligence algorithm to analyze the clinical data of Yiyang central hospital.Based on the data analysis results,we use support vector machine(SVM)and Bayesian network algorithm to establish the prediction model of treatment days,treatment methods and treatment drug analysis.Results:(1)in the original database,there were 7567 patients’ basic information,109302 western medicine orders,9812 traditional medicine orders,8123 prescription orders,3561 traditional Chinese medicine treatments,6812 other treatments and 861 operations in the database of his DB2;26014 patients’ diagnoses,61083 complications,7392 traditional Chinese medicine diagnoses and 15 drug allergy histories in the database of Oracle There are 3 items,87103 items of inspection information of LIS Oracle database,42067 items of inspection information of PACS Oracle database,which are preprocessed with pandas to generate training samples of specific format.(2)In this study,there were 7567 patients,with slightly more female patients,accounting for 51.7% of the total data,and fewer male patients,accounting for 48.3% of the total data;45-75 age group had the largest number of patients,accounting for 73.21% of the total,which indicated that 45-75-year-old patients had the most demand for traditional Chinese medicine treatment,which was also in line with epidemiology;the treatment days were mainly distributed in the range of 5-25 days,and none of them exceeded the total The treatment limit of 30 days,at the same time,there is less than 1 day of treatment;there are 29 kinds of traditional Chinese medicine treatment methods,of which the most widely used is the hot application of traditional Chinese medicine ointment,followed by general acupuncture;there are more than 40 kinds of combination of Chinese patent medicine,including Bailemian capsule and puliku capsule,Huoluo Xiaotong capsule,qianglidingxuan capsule,Yinao capsule and other Chinese patent medicines are used more frequently;according to the classification of diseases,the proportion of cancer is the largest,the proportion of diabetes is also large,and the proportion of coronary heart disease,brain atrophy,spinal and cervical lumbar diseases is also more than 5%,which is also in line with the statistical data of actual diagnosis and treatment.(3)SPSS24.0 was used to establish the knowledge database of traditional Chinese medicine in Yiyang central hospital.(4)Based on the open source language python,the paper constructs a Chinese medicine knowledge data mining platform,which can quickly realize the original cumbersome mining process through a few simple operations.(5)Through the support vector machine(SVM)and Bayesian network algorithm to achieve the prediction of patients’ treatment days,traditional Chinese medicine medication and treatment methods,the feasibility of this algorithm in practical application is verifiedConclusion: This study can help Yiyang central hospital to better grasp the basic information of patients,and provide data support for the establishment of case health records for patients with traditional Chinese medicine.In addition,according to the big data analysis technology,according to different ages,genders and diagnosis results,personalized diagnosis and treatment plans can be provided for patients.Atthe same time,personalized diagnosis and treatment plans can make patients fully understand the required drugs,treatment days and other information,so as to take better care of patients,especially elderly patients’ emotions and help them get more scientific Diagnosis and treatment services.Through this study,a complete database of traditional Chinese medicine is also established for the hospital.According to the analysis and research results,a database of disease and medication strategies is established,which lays the foundation for supporting doctors to quickly determine the treatment plan.This is also the significance of this study.