A calibration-free hybrid neural network accurately estimated central aortic pressure waveforms from photoplethysmograms in 4,374 virtual subjects, with mean absolute errors of 1.99-3.17 mmHg.
Does a calibration-free method using photoplethysmogram and a hybrid neural network accurately estimate central aortic pressure waveform compared to gold-standard values in virtual subjects?
A novel calibration-free neural network method can accurately estimate central aortic pressure waveforms from photoplethysmograms in virtual subjects, offering a promising step toward non-invasive clinical application.
Effect estimate: Mean absolute errors 1.99-3.17 mmHg; R2 0.89-0.98; mean differences -0.39-0.12 mmHg
Central aortic pressure waveform (CAPW) provides critical clinical insights for cardiovascular assessment. This study proposed a calibration-free method to estimate CAPW directly from photoplethysmogram by integrating variational mode decomposition with a hybrid neural network combining temporal convolutional network, gated recurrent unit, and self-attention mechanism. Validation on 4374 virtual subjects via subject-level splitting demonstrated excellent accuracy of the estimated indices against gold-standard values, as evidenced by mean absolute errors of 1.99-3.17 mmHg, coefficients of determination of 0.89-0.98, and mean differences of -0.39-0.12 mmHg. This method provides a promising step toward clinically applicable, calibration-free CAPW estimation.
Du et al. (Thu,) reported a other. Calibration-free method integrating variational mode decomposition with TCN-GRU-SAM vs. Gold-standard values was evaluated on Accuracy of estimated indices against gold-standard values (Mean absolute errors 1.99-3.17 mmHg; R2 0.89-0.98; mean differences -0.39-0.12 mmHg). A calibration-free hybrid neural network accurately estimated central aortic pressure waveforms from photoplethysmograms in 4,374 virtual subjects, with mean absolute errors of 1.99-3.17 mmHg.