Coronary artery disease detection using photoplethysmography

Area of Science:

• Biomedical Engineering
• Cardiovascular Diagnostics
• Signal Processing

Background:

• Coronary artery disease (CAD) poses a significant global health burden.
• Current diagnostic methods can be invasive, costly, or inaccessible for widespread screening.
• Photoplethysmography (PPG) offers a potential non-invasive approach for physiological monitoring.

Purpose of the Study:

• To develop and evaluate a non-invasive technique for coronary artery disease (CAD) detection using photoplethysmography (PPG).
• To establish an inexpensive and accessible screening method suitable for home-based monitoring.
• To assess the feasibility of using time-domain PPG signal analysis for CAD classification.

Main Methods:

• Extraction of distinguishing features from the time-domain analysis of the PPG signal and its second derivative.
• Utilizing a Support Vector Machine (SVM) classifier for the binary classification of CAD patients.
• Performance evaluation conducted on patient data sourced from the MIMIC-II dataset (Intensive Care Unit).

Main Results:

• The developed technique achieved a sensitivity of 85% for CAD detection.
• A specificity of 78% was obtained in classifying CAD patients.
• The results demonstrate the potential of PPG signal analysis for identifying individuals with coronary artery disease.

Conclusions:

• Photoplethysmography (PPG) analysis, particularly its time-domain features, can be effectively utilized for non-invasive coronary artery disease (CAD) screening.
• The proposed technique shows promise as an inexpensive and accessible tool for home monitoring and early detection of CAD.
• Further validation and clinical studies are warranted to fully establish the utility of this PPG-based CAD detection method.