Key result
Cuff-less blood pressure estimation using a 16-layer VGGNet achieved high accuracy compared to oscillometric measurement, with correlation coefficients of 0.91 for systolic and 0.89 for diastolic BP.
Why the study?
Existing cuff-less BP monitoring methods rely on manual feature extraction that cannot characterize the complex relationship between physiological signals and BP.
Does a deep convolutional neural network using ECG and PPW signals accurately estimate blood pressure compared to oscillometric techniques in middle-aged and elderly subjects?
Observational (n=89)
Does a deep convolutional neural network using ECG and PPW signals accurately estimate blood pressure compared to oscillometric techniques in middle-aged and elderly subjects?
Effect estimate: r=0.91 (SBP), r=0.89 (DBP)
A deep learning approach using ECG and PPW signals can accurately estimate blood pressure without a cuff, offering a potential novel method for continuous BP monitoring.
No takes yet. Share an insight, caveat, or question.
May support cuffless BP monitoring feasibility; leaves open larger prospective validation before clinical adoption.
Liu et al. (2019) conducted an observational in Blood pressure monitoring (n=89). 16-layer VGGNet for cuff-less BP estimation from ECG and PPW signals vs. Oscillometric technique-based BP was evaluated on Accuracy in BP estimation (correlation coefficient and estimation error) (r=0.91 (SBP), r=0.89 (DBP)). Cuff-less blood pressure estimation using a 16-layer VGGNet achieved high accuracy compared to oscillometric measurement, with correlation coefficients of 0.91 for systolic and 0.89 for diastolic BP.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: