Satin Bowerbird Optimization-Based Classification Model for Heart Disease Prediction Using Deep Learning in E-Healthcare

Abstract

currently, the medical field is most concerned about cardiovascular disease (CVD), which is a chronic and highly fatal condition that accounts for the highest number of global deaths. The number of cases of heart attacks has been steadily increasing across various age groups, except those below 28 years, as reported by the National Crime Records Bureau (NCRB). Wearable sensor devices have become prevalent in the current healthcare scenario. They have enabled real-time monitoring of health records, thus aiding in the early identification of the risk of heart disease. The accurate diagnosis and prediction of cardiovascular disease are vital in providing appropriate treatment to patients by cardiologists. This study aims to develop a model that can accurately predict cardiovascular diseases and thereby reduce the fatality rates associated with them. The Satin Bowerbird optimization algorithm selects the most significant feature, and an enhanced deep-learning model is employed for classification. Here the performance of the proposed work is compared with other methods such as SVM, Decision Tree, Logistic Regression, Random Forest, and Evolutionary Deep Learning. Its effectiveness is evaluated using accuracy, precision, recall, and Fl-score metrics in PYTHON. The results show that the proposed model achieved 90% accuracy, 94% precision, 91.3% recall, and an F1 score of 92.6%.