Effective heart disease prediction system using data mining techniques.

Area of Science:

• Health Informatics
• Artificial Intelligence in Medicine
• Cardiovascular Disease Research

Background:

• Healthcare generates vast datasets with valuable hidden information for decision-making.
• Advanced data mining techniques are crucial for extracting insights and improving healthcare outcomes.

Purpose of the Study:

• To develop an effective heart disease prediction system (EHDPS) using neural networks.
• To predict the risk level of heart disease in patients.
• To establish relationships between medical factors and heart disease patterns.

Main Methods:

• Utilized a neural network approach for heart disease risk prediction.
• Employed a multilayer perceptron neural network with backpropagation as the training algorithm.
• Incorporated 15 key medical parameters including age, sex, blood pressure, cholesterol, and obesity.

Main Results:

• The developed EHDPS effectively predicts the risk level of heart disease.
• The system demonstrates the ability to identify patterns and relationships among medical factors.
• The diagnostic system provides accurate predictions for patient heart disease risk.

Conclusions:

• The EHDPS offers a valuable tool for early detection and risk assessment of heart disease.
• Neural network models can be effectively applied to complex medical data for disease prediction.
• This system enhances decision-making in healthcare by providing predictive insights into heart disease.