Scholarlink

Heart disease is a leading cause of mortality worldwide, underscoring the need for effective predictive tools for early detection and intervention. This study develops and implements a ml-based heart disease prediction using a dataset from Kaggle.com, encompassing features such as age, resting blood pressure, gender, number of major vessels, cholesterol level, chest pain type, old peak, fasting blood sugar, resting ECG, slope of peak exercise, thalach, exang, and thal. Multiple supervised machine learning algorithms—XGBoost, AdaBoost, K-Nearest Neighbors (KNNs), Support Vector Classifier (SVC), Decision Tree, Gradient Boosting, Logistic Regression, and Random Forest—were trained and evaluated to determine the most accurate model. Random Forest achieved the highest accuracy of 100%, demonstrating its potential for reliable heart disease prediction. Beyond model development, a user-friendly web interface was created to enable user registration, administrative authorization, and real-time heart disease risk assessments based on user-provided health parameters. If heart disease is predicted, the application integrates with the Practo platform to recommend qualified cardiologists, allowing users to book appointments directly. The integration of this capability enables easier usage for the end user and bridges the gap between predictive analytics and ultimately actionable medical care. The proposed system shows how machine learning in medical diagnostics helps lower the cost of medical treatment and improve patient outcomes through early detection, and easy access to medical services.

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

Heart Disease Risk Prediction Using Supervised Machine Learning Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Shraddha Pandey,
  • Sonam Gupta,
  • Pradeep Gupta,
  • Akhilesh Verma

摘要

Heart disease is a leading cause of mortality worldwide, underscoring the need for effective predictive tools for early detection and intervention. This study develops and implements a ml-based heart disease prediction using a dataset from Kaggle.com, encompassing features such as age, resting blood pressure, gender, number of major vessels, cholesterol level, chest pain type, old peak, fasting blood sugar, resting ECG, slope of peak exercise, thalach, exang, and thal. Multiple supervised machine learning algorithms—XGBoost, AdaBoost, K-Nearest Neighbors (KNNs), Support Vector Classifier (SVC), Decision Tree, Gradient Boosting, Logistic Regression, and Random Forest—were trained and evaluated to determine the most accurate model. Random Forest achieved the highest accuracy of 100%, demonstrating its potential for reliable heart disease prediction. Beyond model development, a user-friendly web interface was created to enable user registration, administrative authorization, and real-time heart disease risk assessments based on user-provided health parameters. If heart disease is predicted, the application integrates with the Practo platform to recommend qualified cardiologists, allowing users to book appointments directly. The integration of this capability enables easier usage for the end user and bridges the gap between predictive analytics and ultimately actionable medical care. The proposed system shows how machine learning in medical diagnostics helps lower the cost of medical treatment and improve patient outcomes through early detection, and easy access to medical services.