Heart disease prediction using data mining

Heart disease remains a major cause of mortality worldwide, emphasizing the importance of early detection for effective treatment. This project presents a heart disease prediction system leveraging Multilayer Perceptron (MLP) neural networks to analyze key patient data such as age, cholesterol levels, and blood pressure. The model achieved an accuracy of 85%, balancing precision and recall effectively. A user-friendly interface was developed using Streamlet, enabling healthcare providers to input patient information and receive real-time predictions, streamlining the diagnostic process. Feature importance analysis identified critical factors contributing to heart disease risk, providing valuable insights for personalized care and helping prioritize patient management. The system has undergone extensive validation, demonstrating its reliability in clinical settings. Future enhancements will focus on integrating real-time data from wearable devices to further improve prediction accuracy and expanding the model to address other cardiovascular conditions. By combining artificial intelligence and a user centric design, this system aims to enhance early diagnosis and improve patient outcomes.