Prediction of Coronary Artery Disease Using Machine Learning Techniques with Iris Analysis.

Area of Science:

• Ophthalmology
• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Coronary Artery Disease (CAD) is a leading cause of mortality due to narrowed coronary vessels restricting heart blood flow.
• Early diagnosis and treatment of CAD are crucial for improving patient prognosis and preventing disease progression.
• Current diagnostic methods for CAD can be invasive or require specialized equipment.

Purpose of the Study:

• To propose and evaluate a novel, non-invasive method for diagnosing Coronary Artery Disease (CAD) using iris image analysis.
• To investigate the efficacy of combining iridology with advanced image processing techniques for CAD detection.
• To establish a preliminary diagnostic tool for CAD that complements existing medical imaging and testing modalities.

Main Methods:

• Iris images from 198 volunteers (94 with CAD, 104 without) were analyzed.
• Iris images were processed using integral differential operator and rubber sheet methods, with the heart region cropped.
• Feature extraction involved wavelet transform, statistical analysis, GLCM, and GLRLM, followed by Support Vector Machine (SVM) classification.

Main Results:

• The proposed method achieved a 93% accuracy rate in predicting Coronary Artery Disease (CAD).
• The model demonstrated high performance across various evaluation metrics including sensitivity, specificity, and AUC.
• Preliminary diagnosis of CAD was achieved through iris analysis, potentially reducing the need for traditional tests.

Conclusions:

• Iris image analysis offers a promising non-invasive approach for the preliminary diagnosis of Coronary Artery Disease (CAD).
• This method can potentially serve as a valuable tool in telemedicine systems for remote CAD screening and diagnosis.
• Further research can explore integration into broader telemedicine platforms for accessible cardiovascular health monitoring.