Key result
Using two PWV values cuts BP prediction errors by ~91% versus single-PWV models.
Why the study?
Cuffless BP monitoring techniques based on PWV or PTT often suffer from large errors because they typically rely on a single PWV value, ignoring nonlinear temporal variations across the cardiac cycle.
Does a blood pressure prediction model incorporating time-varying PWV improve prediction accuracy compared to a single PWV model in simulated arteries?
Does a blood pressure prediction model incorporating time-varying PWV improve prediction accuracy compared to a single PWV model in simulated arteries?
Effect estimate: 90.6% and 96.8% reduction in error
Incorporating time-varying pulse wave velocity (diastolic and systolic) into cuffless blood pressure prediction models significantly reduces prediction errors compared to traditional single-PWV methods.
PURPOSE: Continuous, cuffless blood pressure (BP) monitoring devices based on measuring pulse wave velocity (PWV) or pulse transit time (PTT) are emerging but are often plagued by large prediction errors. A key issue is that these techniques typically rely on a single PWV value, assuming a linear response and small arterial wall deformations. However, arterial response to BP is inherently nonlinear, with PWV varying over time [PWV(t)] by up to 50% during a cardiac cycle. This study evaluates the impact of assuming a single PWV on BP prediction accuracy. METHOD: Using a Fluid-structure Interaction (FSI) testbed, we simulate the radial and common carotid arteries with the Holzapfel-Gasser-Ogden (HGO) constitutive model to capture nonlinear arterial behavior under a pulsatile physiological blood flow. Pressure data from FSI simulation are used as the ground truth, while inner area A(t) and two PWV values, at diastole and systole, serve as inputs to BP prediction models. Two models are tested: one using a single PWV value, emulating existing PWV-based BP prediction methods; another using the two PWV values to account for PWV(t). RESULTS: The single-PWV BP model produced prediction errors of 17.44 mmHg and 6.57 mmHg for the radial and carotid arteries, respectively. The model incorporating two PWV values reduced these errors by 90.6% and 96.8%, respectively. CONCLUSION: Relying on a single PWV in BP prediction models can lead to significant errors. To improve BP accuracy, future efforts should focus on incorporating PWV(t), or at least both diastolic and systolic PWV values, into these models.
No takes yet. Share an insight, caveat, or question.
Bantwal et al. (2025) studied Blood pressure prediction. Time-varying PWV model (two PWV values) vs. Single-PWV model was evaluated on Blood pressure prediction error (90.6% and 96.8% reduction in error). Incorporating two pulse wave velocity values to account for temporal variations reduced blood pressure prediction errors by 90.6% in the radial artery and 96.8% in the carotid artery compared to a single-PWV model.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: