The feasibility of early detecting coronary artery disease using deep learning-based algorithm based on...

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Imaging Analysis

Background:

• Coronary Artery Disease (CAD) is a leading cause of death and illness.
• Early-stage CAD is often asymptomatic, leading to delayed diagnosis.
• Current diagnostic methods can be invasive or costly.

Purpose of the Study:

• To develop an artificial intelligence (AI) model for early Coronary Artery Disease (CAD) detection.
• To utilize electrocardiogram (ECG) data exclusively for CAD patient identification.
• To establish a noninvasive and cost-effective screening tool.

Main Methods:

• A convolutional neural network (CNN) model was developed using patient ECG and coronary computed tomography angiography (cCTA) data.
• Data was randomly allocated into training, validation, and testing datasets.
• Model performance was evaluated using accuracy, sensitivity, specificity, PPV, NPV, and AUC.

Main Results:

• The AI model achieved an Area Under the Curve (AUC) of 0.75 for CAD detection.
• The model demonstrated 70.0% accuracy, 68.7% sensitivity, and 70.9% specificity.
• Positive predictive value was 61.2% and negative predictive value was 77.2%.

Conclusions:

• A CNN model trained on ECG data can effectively assist in early Coronary Artery Disease detection.
• This AI-driven ECG analysis offers an efficient, low-cost, and noninvasive diagnostic aid.
• The findings support the use of AI-powered ECG for broader CAD screening.