What makes PPG sensors vulnerable to motion artifacts during workouts?

Photoplethysmography (PPG) is a non-invasive optical technique used to measure blood-volume pulses and derive vital signs by illuminating the skin and detecting changes in light absorption ^[1]. During workouts, wrist-worn PPG sensors are highly vulnerable to motion artifacts ^[2] because user movement alters the tissue-contact-sensor stack and creates baseline drift alongside time-varying system parameters ^[3]. Physical activities such as walking and running cause distortions ^[4] that compromise the reliability of parameters like heart rate ^[5], oxygen saturation, blood pressure, and respiration ^[6].

Current mitigation techniques include advanced signal processing and hardware-level adjustments:
* Adaptive Noise Cancellation (ANC): Algorithms utilizing methods such as the least mean squares (LMS)-Newton algorithm compensate for preliminary motion artifacts ^[7].
* Frequency Estimation and Filtering: Techniques combine heart rate frequency estimation with notch filters to restore the quasi-periodic PPG signal ^[8] and use band-pass filters to cancel DC-current drift ^[9].
* Convolutional Neural Networks: One-dimensional convolutional neural networks (1-D-CNNs) can classify continuous raw PPG segments into clean or artifact-affected data without needing complex manual feature engineering ^[10].
* Dual-Channel and Multi-Wavelength Hardware: Readout systems can incorporate an additional reference channel—such as a secondary organic photodetector channel—to effectively cancel DC drifts and ambient lighting interference without needing auxiliary motion sensors ^[11]. Hardware configurations also explore specific light source wavelengths, brightness levels, and multi-site arrays to optimize performance ^[12].

Would you also like to know how different LED wavelengths affect PPG signal quality?DemanderRéponse terminée. Une option de suivi est disponible.