Enhanced extremely boosted neural network (EXBNet) for effective Heart Disease Prediction

Cardiovascular disease is considered to be one of the major causes of deaths around the world. Hence, a reliable, accurate, and efficient approach is needed to diagnose such diseases at an early stage so as to start the relevant treatment procedure. Machine learning and Neural network plays a critical role in processing massive amount of data in the field of healthcare. Artificial Neural Network (ANN) and Deep Neural Network (DNN) techniques have been used in earlier research to predict heart diseases. The proposed methodology presents an effective heart disease prediction using an Enhanced Extremely boosted neural network (EXBNet) that can assist medical professionals. The proposed method, Enhanced XBNET is a variation of XBNET where Gini Index is used to compute the feature importance of Gradient Descent trees. XBNET is an ensemble model that combines both Gradient Boost tree-based models and Feed Forward Neural networks. The experiments were conducted on the data that has been collected using Kaggle datasets. The performance of proposed methodology is compared against various state-of-art methods using different classifiers and are evaluated using the performance metrics Accuracy, Recall, Precision, and F1 score. The research methodology provides an enhanced performance with an accuracy of 100%. The proposed methodology is compared with various state of art algorithms and it has been proved that the proposed method outperforms other algorithms.