A Hybridized Model for the Prediction of Heart Disease using ML Algorithms
T Penchala Naidu, K Amar Gopal, Sk Rameez Ahmed, R. Revathi, Sk Hasane Ahammad, V. Rajesh, Syed Inthiyaz, K. Saikumar
Abstract
Heart disease is a chronic illness that can affect the circulatory system of the body. It can cause various types of complications such as heart failure and stroke. Machine learning is a promising technology for identifying people with heart disease. Precise and exactness in heart disease diagnosis is playing crucial role for the restraint and therapy of heart failure. Through the conventional way of analysing heart disease, the therapeutic record has judged which isn’t consistent in numerous characteristics. Further, the classification of healthy, infected people, a non-invasive-based procedures that include artificial intelligence (AI) and machine learning (ML) techniques which obey reliability and efficient outcomes. A machine-learning-based analysis approach for heart disease prediction is proposed in this study employing heart disease dataset. Moreover, the proposed system can effortlessly distinguish in addition to categorize individuals with heart disease as of healthy persons. In this study, dataset of Cleveland heart disease with ECG images for the development of hybridised model and then extracting important features using Genetic Algorithm and PSO algorithm and formerly pertaining neural network algorithm to build prediction model and then prediction model will be applied on test data to calculate metrics like prediction accuracy. Subsequently, the proposed machine learning-based methodology under the process of decision assistance enforce aid the medical practitioner to effectively diagnosis the heart patients.