Predictive Analytics Model for Heart Disease: Leveraging Machine Learning Techniques

Present-day common disease is the heart disease that is chronic and dangerous for human being as it leads to death of a person. It accumulates its symptoms continuously from a particular lifestyle since long time that causes heart failure and death. The amount of blood required for the body could not be supported by the heart due to blocked arteries developed in the heart chamber. The model will be able to accurately predict the chances of developing heart disease by analyzing various risk factors and medical data. This study proposes a predictive model for heart disease, utilizing Machine Learning algorithms to detect and mitigate heart diseases early on. With heart disease posing a significant health concern and treatment costs often beyond common means, this system aims to enhance treatment efficacy. By analyzing patient profiles of an individual, the system excels in identifying potential heart problems. The underutilization of healthcare data in the biomedical field prompts the introduction of this system, contributing to cost reduction and improving the quality of heart patient treatment. The XGBoost algorithm plays a crucial role in early detection, as confirmed by a performance assessment utilizing a confusion matrix, demonstrating high accuracy of 93.24%, sensitivity of 93.4%, specificity of 92.9% and F1 score of 93.8%. Following XGBoost model, the Random Forest (RF) model also demonstrates robust performance, with high scores in precision and specificity. These results suggest that ensemble methods such as XGBoost and Random Forest are well-suited for the heart disease prediction task handling complex interactions and nonlinear data. The overall outcome signifies the system's superior performance, promising significant advancements in early detection and treatment of heart diseases.