Effective diagnosis of coronary artery disease using the rotation forest ensemble method

Area of Science:

• Cardiology
• Computational Medicine
• Machine Learning

Background:

• Coronary Artery Disease (CAD) is a leading cause of mortality, necessitating early diagnosis.
• Accurate diagnosis is crucial, especially in early or asymptomatic stages.
• Existing computational methods for CAD diagnosis have limitations in accuracy.

Purpose of the Study:

• To develop and evaluate a novel computational tool for effective Coronary Artery Disease diagnosis.
• To enhance clinical decision-making through improved diagnostic accuracy.
• To compare the proposed method's performance against other classifiers.

Main Methods:

• Implementation of a Rotation Forest ensemble algorithm.
• Utilization of Artificial Neural Networks with Levenberg-Marquardt back propagation as base classifiers.
• Validation using patient data for Coronary Artery Disease diagnosis.

Main Results:

• Achieved a diagnostic accuracy of 91.2% for Coronary Artery Disease.
• Outperformed previous computational methods using the same dataset.
• Demonstrated superior performance compared to other classifiers in diagnosis.

Conclusions:

• The proposed Rotation Forest-based computational tool significantly improves Coronary Artery Disease diagnosis accuracy.
• This method offers a promising approach for supporting clinical decision-making in cardiology.
• The findings represent a new benchmark for computational CAD diagnosis.