Key result
The BioTranslator deep learning framework identified 92.8% of R-peaks from wrist-worn PPG signals during physical activity, achieving a mean absolute error of 51±6.3ms compared to reference ECG.
Why the study?
Wrist-worn photoplethysmography signal quality suffers from motion artifacts in ambulatory settings, making vital parameter estimation challenging.
Does the BioTranslator deep learning framework accurately infer R-peaks and instantaneous heart rate from wrist-worn PPG signals during physical activity compared to reference ECG?
Does the BioTranslator deep learning framework accurately infer R-peaks and instantaneous heart rate from wrist-worn PPG signals during physical activity compared to reference ECG?
Mean Difference: 51
The BioTranslator deep learning framework can effectively translate ambulatory wrist-worn PPG signals into ECG-like signals to accurately infer R-peaks and instantaneous heart rate during physical activity.
May enable PPG-based R-peak detection during activity; hypothesis-generating for clinical use pending larger validation.
Advancements in wireless sensor networks (WSN) technology and miniaturization of wearable sensors have enabled long-term continuous pervasive biomedical signal monitoring. Wrist-worn photoplethysmography (PPG) sensors have gained popularity given their form factor. However the signal quality suffers due to motion artifacts when used in ambulatory settings, making vital parameter estimation a challenging task. In this paper, we present a novel deep learning framework, BioTranslator, for computing the instantaneous heart rate (IHR), using wrist-worn PPG signals collected during physical activity. Using one-dimensional Convolution-Deconvolution Network, we translate a single channel PPG signal to an electrocardiogram(ECG)-like time series signal, from which relevant R-peak information can be inferred enabling IHR measures. The proposed network configuration was evaluated on 12 subjects of the TROIKA dataset, involved in physical activity. The proposed network identifies 92.8% of R-peaks, besides achieving a mean absolute error of 51±6.3ms with respect to reference ECG-derived IHR.
No takes yet. Share an insight, caveat, or question.
Everson et al. (2019) studied Physical activity (n=12). BioTranslator deep learning framework vs. Reference ECG-derived IHR was evaluated on R-peak identification and instantaneous heart rate (IHR) mean absolute error (mean absolute error of 51±6.3ms). The BioTranslator deep learning framework identified 92.8% of R-peaks from wrist-worn PPG signals during physical activity, achieving a mean absolute error of 51±6.3ms compared to reference ECG.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: