Coronary Artery Disease Detection Based on a Novel Multi-Modal Deep-Coding Method Using ECG and PCG Signals

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence in Medicine

Background:

• Coronary artery disease (CAD) is a leading cause of mortality, necessitating accurate and early diagnosis.
• Current diagnostic methods often rely on single-modal data (ECG or PCG), leading to information shortages and reduced precision.
• A multi-modal approach integrating diverse physiological signals is crucial for comprehensive CAD assessment.

Purpose of the Study:

• To develop and evaluate a novel multi-modal deep-learning method for enhanced coronary artery disease (CAD) detection.
• To integrate electrocardiogram (ECG) and phonocardiogram (PCG) signals for improved diagnostic accuracy.
• To address the limitations of single-modal diagnostic approaches in CAD detection.

Main Methods:

• A novel multi-modal learning framework was developed, integrating ECG and PCG signals.
• A deconvolution operation was used to create an ECG-PCG coupling signal, enriching diagnostic information.
• A parallel Convolutional Neural Network (CNN) architecture encoded multi-modal features, followed by an autoencoder for feature compression.
• Support Vector Machine (SVM) classifier was employed for final CAD classification using optimal multi-modal features.

Main Results:

• The proposed multi-modal method achieved high performance on a dataset of 199 subjects.
• Accuracy, sensitivity, specificity, and F1-score reached 98.49%, 98.57%, 98.57%, and 98.89%, respectively.
• The multi-modal approach demonstrated superiority over single-modal methods in CAD detection.

Conclusions:

• The developed multi-modal deep-coding method effectively overcomes information shortages inherent in single-modal signals for CAD detection.
• This approach offers enhanced precision and outperforms existing models in identifying coronary artery disease.
• The study highlights the potential of integrating multi-modal physiological data for advancing CAD diagnosis and patient care.