Hybrid Machine Learning Strategies for Heart Disease Prediction in Healthcare

Abstract

Heart disease is one of the most fundamental human sicknesses on the earth and influences human life badly, and on-time treatment is a must for preventing heart disease. The persistence of heart disease via conventional medical reports has been considered, and it is not applicable in numerous viewpoints. In recent years, the popular strategies in machine learning are used for binary classification of heart disease (healthy, diseased), and these are efficient and reliable. ML has been demonstrated to be compelling in helping with settling on choices and forecasts from the vast amount of information created by the health care industry. We have likewise observed ML methods being utilized in ongoing improvements in various regions of the IoT. Different investigations give just a look into foreseeing heart disease with ML procedures. Though the usage of ML algorithms is reliable, we can still improve the accuracy of these algorithms for efficient heart disease prediction by applying hybrid machine learning algorithms. Those hybrid algorithms find significant features by using machine learning strategies bringing about improving the accuracy in the prediction of heart disease. This model predicts various combinations of features and a few known classification strategies. Moreover, there is an absence of powerful tools to extract the patterns and to discover the hidden patterns in the input data. Automated tools are needed to detect heart disease that will increase medical efficiency and diminishes the cost. In this paper, we have presented a comprehensive review of the conventional techniques used along with the recent research works related to hybrid machine learning methods to detect heart disease.