PPGFeat: a novel MATLAB toolbox for extracting PPG fiducial points

Area of Science:

• Biomedical Engineering
• Signal Processing
• Physiological Monitoring

Background:

• Photoplethysmography (PPG) is a non-invasive optical technique to measure blood volume changes.
• Extracting accurate physiological features from PPG signals is crucial but challenging.
• Existing methods for PPG analysis often require complex feature extraction.

Purpose of the Study:

• To introduce PPGFeat, a novel MATLAB toolbox for analyzing raw photoplethysmography data.
• To provide tools for preprocessing, feature extraction, and fiducial point detection in PPG signals.
• To enhance the accuracy and efficiency of PPG signal analysis for researchers.

Main Methods:

• Developed PPGFeat, a MATLAB toolbox with a graphical user interface.
• Implemented preprocessing techniques: filtering, smoothing, and baseline drift removal.
• Integrated algorithms for detecting and adjusting PPG fiducial points.

Main Results:

• PPGFeat achieved 99% accuracy in identifying fiducial points on the PPG-BP dataset.
• 3038 out of 3066 fiducial points were correctly identified.
• The toolbox significantly reduces errors in fiducial point identification.

Conclusions:

• PPGFeat offers a valuable resource for researchers analyzing photoplethysmography signals.
• The toolbox enhances the reliability of vital sign measurement and disease risk assessment.
• Accurate feature extraction from PPG data is improved with PPGFeat.