Scholarlink

<p>Cardiovascular disease (CVD) remains a significant global health concern, necessitating early detection and accurate prediction for effective intervention. Machine learning (ML) offers a data-driven approach to analyzing patient data, identifying complex patterns, and predicting CVD risk factors such as blood pressure, cholesterol, and genetic predispositions. This study employs the Heart Disease UCI dataset, comprising 303 cases and 14 attributes, to develop predictive ML models. A comprehensive feature engineering approach was applied, including Chi-Square feature selection, where the 10 most influential features were identified to enhance model performance. After feature selection, hyperparameter tuning was performed to optimize five classifiers: Random Forest, K-Nearest Neighbors, Decision Tree, Support Vector Machine, and Logistic Regression. The results indicate significant improvements in accuracy after feature engineering and fine-tuning, with the Random Forest classifier achieving the highest accuracy of 93.41%, outperforming previous work on the same dataset by 3.41%, followed by SVM at 91.21%. These findings underscore the potential of ML in early CVD prediction, emphasizing its role in proactive risk management and healthcare. The integration of ML-driven CVD risk assessment paves the way for personalized healthcare, benefiting both patients and medical professionals by facilitating early intervention, improving patient outcomes, and optimizing healthcare resource allocation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

HPML-CVD: Hyperparameter Tuned Machine Learning Model to Predict Cardiovascular Disease

  • 期刊论文

  • 发表日期：

    2025年4月7日

查看全文
  • Nazim Uddin,
  • Md. Sahadat Hossain,
  • Md. Zulfiker Mahmud,
  • Rebina Ferdous,
  • Md. Alamin Talukder

摘要

Cardiovascular disease (CVD) remains a significant global health concern, necessitating early detection and accurate prediction for effective intervention. Machine learning (ML) offers a data-driven approach to analyzing patient data, identifying complex patterns, and predicting CVD risk factors such as blood pressure, cholesterol, and genetic predispositions. This study employs the Heart Disease UCI dataset, comprising 303 cases and 14 attributes, to develop predictive ML models. A comprehensive feature engineering approach was applied, including Chi-Square feature selection, where the 10 most influential features were identified to enhance model performance. After feature selection, hyperparameter tuning was performed to optimize five classifiers: Random Forest, K-Nearest Neighbors, Decision Tree, Support Vector Machine, and Logistic Regression. The results indicate significant improvements in accuracy after feature engineering and fine-tuning, with the Random Forest classifier achieving the highest accuracy of 93.41%, outperforming previous work on the same dataset by 3.41%, followed by SVM at 91.21%. These findings underscore the potential of ML in early CVD prediction, emphasizing its role in proactive risk management and healthcare. The integration of ML-driven CVD risk assessment paves the way for personalized healthcare, benefiting both patients and medical professionals by facilitating early intervention, improving patient outcomes, and optimizing healthcare resource allocation.