Vascular waveform analysis using Bayesian pulse deconvolution
Preprint
Abstract
Vascular waveforms, which measure bulk flow in blood vessels, are widely used to measure vital signs, diagnose conditions, and predict long-term health outcomes. Analyzing vascular waveforms depends on three fundamentally interdependent tasks: signal filtering, pulse timing detection, and pulse shape extraction. We hypothesized that Bayesian pulse deconvolution can achieve improved performance on all three tasks by solving them jointly. This method uses an analytical, generative model of vascular waveforms with priors informed by physical and biological domain knowledge. In simulations, Bayesian pulse deconvolution achieves better performance on all tasks compared with existing algorithms: 90% reduction of median filtering error, 60% reduction in pulse timing error, and 85% reduction in shape extraction error. The advantages in simulations extend to human recordings of photoplethysmography waveforms. Taking real time-synchronized electrocardiogram R-R intervals as a proxy ground truth, Bayesian pulse deconvolution achieves 40% lower pulse interval estimation error (RMSE = 5.1 ms) compared with typical algorithms (RMSE = 8.3 ms, p=1e–10). By extracting more accurate and informative insights from vascular waveforms, Bayesian pulse deconvolution could advance a wide array of health technologies that rely on interpreting signals from blood vessels.Keywords
Funding Information
- National Science Foundation (BCS-1932619)
- National Science Foundation (2205084)
- Office of Naval Research (N00014-22-1-2110)
- National Institutes of Health (F31HL176112)
- National Institutes of Health (U01NS131914)
- National Institutes of Health (1U01DK140939)
- Wu Tsai Human Performance Alliance at Stanford University
- Wu Tsai Neurosciences Institute at Stanford University
- Stanford Bio-X Interdisciplinary Biosciences Institute
- Stanford Institute for Human-Centered Artificial Intelligence, Stanford University
- Chan Zuckerberg Biohub
This publication has 57 references indexed in Scilit:
- Cuffless Blood Pressure Measurement Devices—International Perspectives on Accuracy and Clinical UseJAMA Cardiology, 2025
- Interpretable deep learning for deconvolutional analysis of neural signalsNeuron, 2025
- Validating cuffless continuous blood pressure monitoring devicesCardiovascular Digital Health Journal, 2023
- Effects of noise and filtering strategies on the extraction of pulse rate variability from photoplethysmogramsBiomedical Signal Processing and Control, 2022
- Heart-rate tuned comb filters for processing photoplethysmogram (PPG) signals in pulse oximetryJournal of Clinical Monitoring and Computing, 2020
- Verification of the coupled‐momentum method with Womersley's Deformable Wall analytical solutionInternational Journal for Numerical Methods in Biomedical Engineering, 2019
- Signal quality measure for pulsatile physiological signals using morphological features: Applications in reliability measure for pulse oximetryInformatics in Medicine Unlocked, 2019
- Modeling of the photoplethysmogram during atrial fibrillationComputers in Biology and Medicine, 2017
- Cardiorespiratory fitness estimation in free-living using wearable sensorsArtificial Intelligence in Medicine, 2016
- Evaluation of filtering methods for acquiring radial intra-artery blood pressure waveformsJournal of Clinical Monitoring and Computing, 2014