Research And Development Of Cardiovascular And Cerebrovascular Diseases Diagnosis System Based On NB And IGA Algorithm

According to statistics,cardiovascular and cerebrovascular diseases are the chronic diseases with the highest mortality and the highest mortality rate among Chinese residents,which seriously affect the health of Chinese residents.Early detection and early prevention of the disease can not only save the medical expenses of residents,but also improve the health level of Chinese residents.Therefore,in order to improve the diagnostic efficiency and accuracy of cardiovascular and cerebrovascular diseases by medical staff,this paper focuses on the application of related feature attribute selection algorithms and classification algorithms in medical health diagnosis system,and designs and develops diagnosis system of cardiovascular and cerebrovascular diseases based on Naive Bayes Algorithm and Improved Genetic Algorithm as well.The system builds a diagnostic model of cardiovascular and cerebrovascular diseases,and presents the corresponding results according to the examination data of patients,so as to provide references for medical staff and assist them to make decisions on cardiovascular and cerebrovascular diseases..My main work is as follows:(1)A large number of domestic and foreign research papers have been reviewed,and the current research status of medical diagnosis system has been emphatically studied.At the same time,the chronic diseases that have the greatest impact on the health status of Chinese residents have been also studied,and the research direction and purpose of this paper have been established.(2)Using genetic algorithm to select the related attributes of cardiovascular and cerebrovascular diseases,aiming at the problems of local convergence,low operational efficiency,etc.This paper proposes a hybrid algorithm based on CFS and improved genetic algorithm(CFS-IGA)for attribute selection,which improves the fitness function of Genetic Algorithm by attribute correlation selection(CFS),so that the selected attribute can meet the requirement of attribute independence on classification by Naive Bayesian Classifier;which adopts parallel and improved elite retention strategy to improve the problems of low running efficiency and low population fitness of Genetic Algorithm.The global search ability of genetic algorithm is improved by introducing adaptive crossover,mutation operation and introducing the degree of acquaintance in the crossover operator.Finally,compared the experimental results of the attribute selection with the attribute correlation selection(CFS),the standard genetic algorithm(GA),and Improved Genetic Algorithm based on the information gain(GA-IG),the fusion algorithm(CFS-IGA)of CFS and Improved Genetic Algorithm.Experiments show that CFS-IGA has a better effect on the dimension and classification accuracy of attribute subsets,which can effectively remove redundant and irrelevant attributes,and select the attribute subset suitable for prediction of cardiovascular and cerebrovascular diseases.(3)The diagnostic model for cardiovascular and cerebrovascular diseases has been constructed by combining Naive Bayesian Classification Algorithm with Decision Tree Classification Algorithm C5.0.Firstly,for the problem that the Naive Bayesian Algorithm has difficulty in probability estimation,the optimization function is adopted to improve the Naive Bayes Algorithm.For the problem of over-fitting in Decision Tree Algorithm,the pre-pruning method is used to improve the algorithm and the probability is used to evaluate the category of leaf nodes.Then,a diagnostic model for cardiovascular and cerebrovascular diseases has been constructed by combining two improved algorithms(B-RNB-C5.0).Finally,by comparing the experimental results of classification models from B-RNB-C5.0,RNB,NB and C5.0,it has been verified that disease diagnostic model based on B-RNB-C5.0 can play a very good diagnostic effect and is suitable for the diagnosis of cardiovascular and cerebrovascular diseases.(4)According to the design requirements of cardiovascular and cerebrovascular disease diagnosis system,a diagnosis system based on Naive Bayes Algorithm and Improved Genetic Algorithm has been designed,and the function of the various modules of the system has been developed and tested.It shows that the function of this system achieves the expected goal and realizes the auxiliary diagnosis of cardiovascular and cerebrovascular diseases.