Online ISSN: 2515-8260

Keywords : heart disease, association, Linear Regression Model, principal component analysis, Decision tree


Prediction Of Heart Disease Using Hybrid Linear Regression

K. Srinivas; B. Kavitha Rani; M. Vara Prasad Rao; Raj Kumar Patra; G. Madhukar; A. Mahendar

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 5, Pages 1159-1171

Heart disease (HD) is one of the most common diseases, and early diagnosis of this disease is a vital activity for many health care providers to avoid and save lives for their patients. Heart disease accounts to be the leading cause of death across the globe. Health sector contains hidden information which helps in making early decisions by predicting existing disease such as coronary heart disease using machine learning methods. The proposed Hybrid Linear Regression Model (HLRM) implemented in two phases. Initially, data preprocessing is done; missing values are imputed with KNN and simple mean imputation and next Principal Component Analysis is used to extract the most contributing attributes for the cause of disease. Second, Stochastic Gradient Descent is the linear regression used to record the probability values of dependent variables, in order to determine the relationship between the dependent and independent variables. The overall prediction accuracy of the proposed model is observed as 89.13%. The outcome of this study will help as a reference for medical practitioners and also as a research platform for the academia