Key result
Adaptive filtering and multi-resolution decomposition techniques are optimal for motion artifact removal, while machine learning-based approaches are the future perspective for heart rate tracking.
Why the study?
PPG-based heart rate tracking is challenging because motion artifacts mask the location of heart rate peaks in spectra, requiring effective removal and tracking techniques.
This review highlights that adaptive filtering and multi-resolution decomposition are optimal for motion artifact removal in PPG, while machine learning approaches hold promise for heart rate tracking.
PPG heart rate tracking remains hindered by motion artifacts; leaves open need for robust removal methods before reliable clinical deployment.
Non-invasive photoplethysmography (PPG) technology was developed to track heart rate during motion. Automated analysis of PPG has made it useful in both clinical and non-clinical applications. However, PPG-based heart rate tracking is a challenging problem due to motion artifacts (MAs) which are main contributors towards signal degradation as they mask the location of heart rate peak in the spectra. A practical analysis system must have good performance in MA removal as well as in tracking. In this article, we have presented state-of-art techniques in both areas of the automated analysis, i.e., MA removal and heart rate tracking, and have concluded that adaptive filtering and multi-resolution decomposition techniques are better for MA removal and machine learning-based approaches are future perspective of heart rate tracking. Hence, future systems will be composed of machine learning-based trackers fed with either empirically decomposed signal or from output of adaptive filter.
No takes yet. Share an insight, caveat, or question.
Ismail et al. (2021) conducted a review in Motion artifacts in photoplethysmography (PPG) signals. Automated analysis techniques (MA removal and heart rate tracking) was evaluated. Adaptive filtering and multi-resolution decomposition techniques are optimal for motion artifact removal, while machine learning-based approaches are the future perspective for heart rate tracking.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: