Deep Learning Based Health Data Analysis To Predict Cardiovascular Disease

Computer science and Engineering

One of the most hazardous disease which causes death to humans all over the world since 15 years is the heart disease. The patient can prevent the heart disease if it is predicted at an earlier stage, this will be also useful for the medical practitioners to understand the cause of heart attack and can avoid before its actual occurrence. The better clarification for the issue is determining the patient’s health issue and analyzing the health problems in prospect so that the specialists can begin the treatment to provide better outcome. It is better than stand-in at later when the patient is in danger and prediction of the heart disease is difficult to analyze and is widely researched area. In this heart disease prediction, several research and advance technology are verified. This paper is suggested to provide the detailed description about the advantage of the techniques and prediction model developed for heart disease. To eliminate the problem in heart disease (HD), doctors and many scientists have suggested to utilize intelligent methods i.e., HD prediction problem is solved using artificial intelligence. The fundamental and major factor which causes death is mis-prediction of disease. An intelligent system is designed to prevent the mis-prediction of heart disease. This paper is a simulation which provides better performance than the traditional diagnostic method. In this paper Artificial Neural Network (ANN), which based on the conventional neural networks is suggested for predicting heart disease. The proposed system is a simulation which provides the better performance than the traditional diagnostic method that is used to diagnose the heart disease. The exploitation and exploration of the binary and multi-class heart disease prediction is obtained by Ant Colony Optimization technique. The supervised learning algorithm in neural network provides better performance in classification task.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.