Key result
The PPG2ABP deep learning model estimated continuous arterial blood pressure waveforms from photoplethysmogram signals with a mean absolute error of 4.604 mmHg.
Why the study?
Continuous blood pressure monitoring typically requires invasive procedures, and existing non-invasive photoplethysmogram-based methods rely on handcrafted feature computation from ideally shaped signals.
Effect estimate: MAE 4.604 mmHg
A novel deep learning pipeline, PPG2ABP, can accurately estimate continuous arterial blood pressure waveforms and standard BP metrics directly from non-invasive PPG signals, meeting established accuracy standards for DBP and MAP.
PPG-based ABP waveform prediction may enable noninvasive monitoring; leaves open prospective validation before clinical use.
Cardiovascular diseases are one of the most severe causes of mortality, a heavy toll of lives annually throughout the world. The continuous of blood pressure seems to be the most viable option, but this an invasive process, bringing about several layers of complexities. motivates us to develop a method to predict the continuous arterial blood (ABP) waveform through a non-invasive approach using (PPG) signals. In addition we explore the advantage of deep as it would free us from sticking to ideally shaped PPG signals only, making handcrafted feature computation irrelevant, which is a shortcoming of existing approaches. Thus, we present, PPG2ABP, a deep learning based, that manages to predict the continuous ABP waveform from the input PPG, with a mean absolute error of 4.604 mmHg, preserving the shape, and phase in unison. However, the more astounding success of PPG2ABP out to be that the computed values of DBP, MAP and SBP from the predicted waveform outperforms the existing works under several metrics, despite that2ABP is not explicitly trained to do so.
No takes yet. Share an insight, caveat, or question.
Nabil Ibtehaz (2020) studied Blood pressure monitoring (n=942). PPG2ABP (deep learning model) vs. Ground truth Arterial Blood Pressure (ABP) waveforms was evaluated on Mean absolute error (MAE) of continuous ABP waveform estimation (MAE 4.604 mmHg). The PPG2ABP deep learning model estimated continuous arterial blood pressure waveforms from photoplethysmogram signals with a mean absolute error of 4.604 mmHg.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: