Key result
A novel machine learning framework for instantaneous heart rate monitoring from wrist-ECG achieved a mean absolute error of 1.4 BPM at a -7 dB signal-to-noise ratio.
Why the study?
Does a novel machine learning framework improve instantaneous heart rate monitoring accuracy from motion-artifact-corrupted wrist-ECG signals?
Population
Wrist-ECG dataset acquired by a semi-customized platform and a public dataset
Comparison
A novel machine learning-enabled framework… vs Well-established approaches
Design
Other
Authors
Loading...
May enable motion-robust wrist-ECG monitoring; leaves open prospective clinical validation.
Does a novel machine learning framework improve instantaneous heart rate monitoring accuracy from motion-artifact-corrupted wrist-ECG signals?
A novel machine learning framework can robustly and accurately estimate instantaneous heart rate from wrist-ECG signals heavily corrupted by motion artifacts.
Zhang et al. (2016) studied this question. Machine learning-enabled framework for instantaneous heart rate monitoring vs. Well-established approaches was evaluated on Mean absolute error of estimated instantaneous heart rate at -7 dB signal-to-noise ratio. A novel machine learning framework for instantaneous heart rate monitoring from wrist-ECG achieved a mean absolute error of 1.4 BPM at a -7 dB signal-to-noise ratio.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: