Utilizing Machine Learning Classification for Heart Disease Identification in E-Healthcare: A Comprehensive Approach
Abstract
This research focuses on early detection of heart disease symptoms using patient data and real-time user input. Modern healthcare data includes detailed demographic and symptom information, allowing for comprehensive analysis. Our proposed method utilizes this data for classification, comparing recent healthcare data with baseline distributions. Machine learning techniques such as Logistic Regression, K-Nearest Neighbors, Random Forest, and XGBoost are employed for training and testing. Classifier performance is evaluated to make predictions.
Keywords
Download Options
Introduction
Cardiac arrest is a fatal event caused by sudden coronary thrombosis, resulting in the death of heart muscle. It occurs when oxygen-rich blood flow to the heart is blocked due to plaque buildup in the arteries, a condition known as atherosclerosis. This plaque can rupture, causing a blood clot that obstructs blood flow and leads to heart muscle death. The resulting damage may not be immediately apparent, leading to long-term complications. Most heart attacks are linked to coronary heart disease, where plaque accumulates in the coronary arteries. Early prediction is crucial for diagnosing cardiac issues and preventing fatalities. Severe spasms in coronary arteries can also trigger heart attacks, while complications like heart failure and life-threatening arrhythmias may arise. Factors such as education and lifestyle contribute to an individual's predisposition to cardiac disease, with socioeconomic status playing a significant role.
Conclusion
In this paper, a reliable multi-process Machine Learning method for building a Heart disease risk prediction system is proposed, showing higher accuracy compared to existing systems. Detecting heart disease symptoms early is crucial to reduce mortality rates. The study aims to define effective data mining techniques for heart disease prediction using only 14 essential attributes. Four data mining classification techniques were applied, with K-nearest neighbor and random forest showing the best results. Further research can explore additional data mining techniques like time series and clustering to improve accuracy for early prediction of heart disease.