Heart Disease Detection using Logistic Regression Model

Cardiovascular disease, also known as heart disease, is one of the main causes of death worldwide, accounting for a significant fatal disease known to humans in the current time, not just in India but globally as a whole. Therefore, in order to take the appropriate steps to improve the patient's prospects for their survival, requires a more accurate and trustworthy method for early heart disease detection. Machine learning has been used to automate the analysis of vast volumes of complex data in a range of medical datasets, which is where it becomes essential. The automated detection of cardiovascular problem could be accomplished by offering a substantial sample of previously collected information on individual’s age, blood sugar, blood pressure and more variables to develop solution for an informed judgement and effectively forecast silent heart disorders. Many academics have been using machine learning technologies to help the healthcare industry in recent years by pointing out worries about the possibility of heart disease developing in the future and providing the accurate diagnosis for an extra degree of assurance. In an attempt to develop automated method for prediction of potentially lethal cardiac disorders in patients, in this work algorithm is proposed and evaluated performance of different classifiers, such as support vector machine, random forest and logistic regressions, for effective detection of heart disease. Proposed method combines the advantages of the more advanced and weaker classifiers to accomplish hybrid classifications.