Scholarlink

Early death related to cardiovascular disease and data mining in the Detection of Metabolic Disorders highlights the critical role of predictive analytics in identifying at-risk individuals, enabling early intervention, and improving patient outcomes through the analysis of large-scale health data. This paper proposes a Neural Network-based classifier model in (DSHDPS) Decision Support System for Heart Disease Prediction to improve the prediction accuracy and categorize events as well as Percutaneous Coronary Intervention PCI, Angina, Acute Myocardial Infarction AMI but not least or avoid prominent events will be some [TR28] Coronary Artery Bypass Graft surgery and classifying other heart diseases. The model, combined with Naïve Bayes method, clearly improves the prediction capability and is a strong apparatus to corroborate the impact of risk factors on CVD occurrences. The health care sector is one which has enormous data that simply goes underutilized. The application of data mining in this report that will find some relations among heart disease parameters is most useful work. The study further proves that using predictive modeling and data mining in coordination with a decision support framework for the prediction of heart diseases will make them more accurate. The Neural Network model with a classification accuracy of 90.63% outperformed Decision Tree Classifier which had an accuracy rate of only 86.55%, and the Naïve Bayes Model—having stuck to low estimates, kept its win percentage down at just under 74%. The Neural Network got a lower error rate (11.24% loss), followed by Decision Tree (15.21% loss) and Naïve Bayes (12.47%). In conclusion, the Naïve Bayes-based prediction model is integrated with Neural Network manners to create a very accurate approach. Conclusions: The hybrid method increases the prediction accuracy, providing a critical option for deep analysis of multifaceted risk factors behind heart disease events. DSHDPS can also run an online survey system for anyone, using which helps to determine and predict the risk of heart disease.

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

Prediction and Evaluation of the Levels of Risk Factors Contributing Toward Coronary Heart Disease Using a Machine Learning Approach

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • A. Sheik Abdullah,
  • Utkarsh Mishra,
  • Mohammad Abbas,
  • Shivansh Dhiman,
  • Nunna Eswar Sri Sai Harshith

摘要

Early death related to cardiovascular disease and data mining in the Detection of Metabolic Disorders highlights the critical role of predictive analytics in identifying at-risk individuals, enabling early intervention, and improving patient outcomes through the analysis of large-scale health data. This paper proposes a Neural Network-based classifier model in (DSHDPS) Decision Support System for Heart Disease Prediction to improve the prediction accuracy and categorize events as well as Percutaneous Coronary Intervention PCI, Angina, Acute Myocardial Infarction AMI but not least or avoid prominent events will be some [TR28] Coronary Artery Bypass Graft surgery and classifying other heart diseases. The model, combined with Naïve Bayes method, clearly improves the prediction capability and is a strong apparatus to corroborate the impact of risk factors on CVD occurrences. The health care sector is one which has enormous data that simply goes underutilized. The application of data mining in this report that will find some relations among heart disease parameters is most useful work. The study further proves that using predictive modeling and data mining in coordination with a decision support framework for the prediction of heart diseases will make them more accurate. The Neural Network model with a classification accuracy of 90.63% outperformed Decision Tree Classifier which had an accuracy rate of only 86.55%, and the Naïve Bayes Model—having stuck to low estimates, kept its win percentage down at just under 74%. The Neural Network got a lower error rate (11.24% loss), followed by Decision Tree (15.21% loss) and Naïve Bayes (12.47%). In conclusion, the Naïve Bayes-based prediction model is integrated with Neural Network manners to create a very accurate approach. Conclusions: The hybrid method increases the prediction accuracy, providing a critical option for deep analysis of multifaceted risk factors behind heart disease events. DSHDPS can also run an online survey system for anyone, using which helps to determine and predict the risk of heart disease.