PPGFeat

PPGFeat implements MATLAB-based analysis of raw photoplethysmography (PPG) waveforms for signal preprocessing, derivative computation, and fiducial point detection to support physiological measurements such as heart rate and blood pressure estimation.

Key Features:

• Preprocessing Techniques: Provides filtering, smoothing, and baseline drift removal for preparing raw PPG signals for analysis.
• Derivative Calculations: Computes derivatives of the PPG waveform to support extraction of physiological waveform features.
• Fiducial Point Detection: Detects and marks fiducial points in PPG signals using dedicated algorithms to reduce errors from manual identification.

Scientific Applications:

• Vital sign estimation: Estimation of heart rate and blood pressure from PPG waveform features.
• Risk stratification: Identification of individuals at increased risk for certain diseases using PPG-derived features.

Methodology:

Preprocessing (filtering, smoothing, baseline drift removal), derivative calculation, and fiducial point detection algorithms were applied; performance was evaluated on the PPG-BP dataset, achieving 99% accuracy with 3038 of 3066 fiducial points correctly identified.