Estimating relative respiratory effort from features of photo-plethysmography signal

The gold standard method for measuring respiratory effort (esophageal pressure measurement) is invasive, uncomfortable, and itself can disrupt sleep. As a consequence, majority of sleep studies use an alternate sensor, typically respiratory bands, which, however, do not measure respiratory effort. Typically they indicate changes in thoracic volume, and so are more a secondary sensor of respiratory movement rather than respiratory effort. In this study, we aim to look at how features extracted from finger Photo Plethysmogram (PPG) signals correlate with changes in esophageal pressure signal. Principle component analysis was used to derive the relative respiratory effort signals using pulse to pulse intervals, pulse wave amplitudes, area of pulse and wavelet decomposed band (0.15~0.4 Hz) of PPG signals.