Research On Predictive Model Of Cardiovascular Disease

Cardiovascular disease is a chronic disease with a high risk of disability and high mortality.In recent years,it has become the biggest obstacle to the development of China’s public medical and health services.The study found that the focus of reducing the number of patients with cardiovascular disease is "pre-diagnosis intervention" rather than "post-diagnosis treatment." Since machine learning methods can obtain high accuracy when processing complex datas,the use of machine learning methods to establish cardiovascular disease prediction models is an important means of "pre-diagnosis intervention".In this context,this article aims to build a cardiovascular disease prediction model based on machine learning methods,starting from three aspects: data processing,mathematical modeling,and system design,and then build a cardiovascular disease prediction system.Accurate prediction of the probability of vascular disease achieves the purpose of early intervention and precise treatment.The main research contents of this article are as follows:(1)Preprocessed the experimental data.From the Heart Disease data set of the UCI machine learning database provided by the University of California,the data attributes suitable for modeling were selected,and the missing values in the data set were filled;in order to solve the problem of non-uniform dimensions,use data normalization to map it to a certain range to eliminate the influence of dimensions;and use one-hot codes to convert text-labeled data into numerical data,improve data quality and establish a training set and a test set.(2)Built a machine learning disease prediction model.Using random forest,support vector machine,and convolutional neural network algorithms to build three cardiovascular disease prediction models,and applying them on the UCI data set,the accuracy rates reached 87.54%,84.52%,and 94.48% respectively.Comprehensive comparison of different models accuracy rate,recall rate,ROC curve,etc.,finally chose the convolutional neural network model as the final cardiovascular disease prediction model.(3)Designed and developed a cardiovascular disease prediction system.Applying computer software system development technology,using Python language,lightweight Web development framework Flask,using html5,css and Java Script technology to design and develop a cardiovascular disease expert early warning system,and embed the cardiovascular disease prediction model based on convolutional neural network.The system realizes the functions of user information recording,online prediction,analysis and suggestion,etc.This article studies the cardiovascular disease prediction model and system based on machine learning,and carries out early intervention before illness,reduces the risk of illness from the source,which can provide reference suggestions for clinical examination.