Key result
XGBoost achieves ~90% accuracy for classifying heart disease, outperforming ten other machine learning models.
Why the study?
The prevalence of heart attacks and heart failure has increased across different age groups, motivating the use of machine learning for predictive modeling.
Answers are grounded in this paper and cite their sources.
Machine learning algorithms, particularly XGBoost, can effectively predict the presence of heart disease using easily obtainable demographic and clinical features.
Machine learning is recognized as one of the fastest-growing technologies and is widely applied across various fields. The healthcare sector benefits significantly from advances in this technology. In recent years, the prevalence of heart attacks and heart failure has increased across different age groups. Heart failure is a condition in which the heart is unable to deliver sufficient blood to meet the body's metabolic demands and may arise from various underlying causes. A dataset comprising 918 records was analyzed, with missing values addressed, outliers handled, and appropriate preprocessing procedures applied. The dataset was then classified using eleven machine learning algorithms, with feature standardization applied selectively to scale-sensitive models. GridSearchCV was used for hyperparameter optimization, and the models were evaluated using standard classification performance metrics.
No takes yet. Share an insight, caveat, or question.
KARAÇAY et al. (2026) studied Heart disease (n=918). Machine Learning Algorithms (XGBoost) vs. Other ML algorithms (e.g., Decision Tree, Naive Bayes, CatBoost) was evaluated on Classification accuracy for predicting heart disease (95% CI 0.8587-0.9402). XGBoost achieved the highest predictive accuracy of 90.22% and an F1-score of 91.18% among eleven evaluated machine learning algorithms for classifying heart disease.
Synapse has enriched 10 related papers. Consider them for comparative context: