Automated Diagnosis of Coronary Artery Disease: A Review and Workflow

Area of Science:

• Cardiology
• Medical Informatics
• Machine Learning

Background:

• Coronary artery disease (CAD) poses significant health risks, including sudden cardiac death.
• Current diagnostic procedures for CAD are often expensive and time-consuming.

Purpose of the Study:

• To review state-of-the-art methods for automatic CAD classification up to 2017.
• To identify optimal methods and classifiers for CAD identification.

Main Methods:

• Proposed two distinct workflows (Parameter Sets A and B) for CAD classification.
• Included preprocessing, feature extraction, feature selection, and classification stages.
• Evaluated Support Vector Machine (SVM) as a key classifier.

Main Results:

• Identified SVM as a promising classifier for CAD detection.
• Highlighted the importance of proper feature extraction from noninvasive signals.
• Established a protocol for future evaluations of automatic CAD diagnosis.

Conclusions:

• Automatic CAD classification methods require rigorous evaluation protocols.
• Further research is needed to optimize feature extraction from noninvasive signals for improved CAD diagnosis.