Research On Auxiliary Diagnosis System Of Stroke Based On Improved Decision Tree Algorithm

Stroke is an acute cerebrovascular disease which has high mortality and disability rate,and it is difficult to fully recover and recover.It will have a negative impact on the patient’s healthy and autonomous life for a long time after the illness.At present,the electronic management system of disease diagnosis and treatment has been established in China,and a large number of medical data such as treatment,treatment and rehabilitation of stroke patients have been accumulated.How to mine the value of these data and establish a diagnosis and treatment auxiliary decision-making system is the important path to improve the medical efficiency and intelligent level at this stage,and it is also the focus of this paper.On the basis of combing and summarizing the research status of domestic and foreign medical data mining,stroke diagnosis and treatment,and auxiliary decision system design,the research work of this paper is as follows:(1)The instance data is grouped by K-Means clustering algorithm,and the training set is randomly sampled to ensure that there are a certain number of instances in each group,and the information of the data set is more accurate,and then iteratively aggregated by K-Means mean Classes,based on their similarity index and distance as the standard,effectively discover K classes in a given data set,and obtain cluster centers on the basis of numerical mean analysis.The verification results show that the accuracy of the decision tree can be significantly improved after the clustering process.(2)An improved decision tree algorithm based on K-Means clustering sampling(IDTACS algorithm)is proposed.Based on K-Means clustering,ID3 decision tree algorithm is used to analyze the clustering data,and the speed of information entropy decline is taken as the standard,and the test attribute standard is selected,at each decision node,the attribute with the highest information gain that has not been selected is selected as the partition standard of the decision tree,until the decision tree is finally generated.Based on the results of data analysis,this article effectively explored the decision tree relationship between high-risk factors such as hypertension,hyperlipidemia,patient age,patient smoking history,patient history of stroke,history of hyper-homocysteinemia,history of diabetes and the incidence of stroke.(3)Based on the analysis of the decision tree improvement algorithm,based on the large amount of first-hand and rich stroke disease data accumulated by the Baise People’s Hospital,relying on the existing database and using the improved decision tree algorithm,analyze the internal correlation in stroke diagnosis and treatment,find the implicit knowledge in stroke data,build an expert knowledge base and establish an auxiliary diagnosis system for stroke diagnosis and treatment,to better improve the accuracy of disease diagnosis and the efficiency of diagnosis and treatment.The stroke auxiliary diagnosis system based on the improved decision tree algorithm is conducive to the early diagnosis of stroke,the support for the diagnosis of primary doctors,and the long-term dynamic monitoring of stroke patients.Most of the disease patients in Baise people’s hospital are ethnic minority residents at the junction of Yunnan,Guizhou and Guangxi provinces,which has strong regional characteristics.It has important guiding and practical significance for the research of ethnic minority stroke data mining and expert system in this region.