CardioAI: An Intelligent Heart Disease Prediction and Detection System

Abstract

Cardiopulmonary disease continues to pose a substantial problem in medicine owing to its elevated incidence and fatality rate. Significant attempts have been undertaken to decrease mortality and alleviate its effects. This paper presents a combined deep learning algorithm aimed at predicting risk for coronary artery by utilizing an extensive array of medical and analytical attributes. The framework incorporates various sophisticated methods, notably k-nearest neighbors (KNN), extreme gradient boosting (XGBoost), long short-term memory networks (LSTM), and convolutional neural networks (CNN), to improve the precision and dependability of cardiac disease forecasting. The model's efficacy was assessed utilizing the Cuyahoga information set, comprising 303 specimens, in conjunction with an aggregated collection of 1,500 specimens obtained from local hospitals in Lucknow, India. Investigations demonstrate that the suggested model attains an estimated reliability of 99.8% on the aggregated dataset, surpassing current methodologies. The results obtained underscore the model's capacity to enhance early identification and treatment results in heart attack risk evaluation.