Trans-stenotic pressure gradient estimation using a modified Bernoulli equation
Ali Amiri, Johan T. Padding, Selene Pirola, Willian Hogendoorn · 2026
Read Trans-stenotic pressure gradient estimation using a modified Bernoulli equation by Ali Amiri, Johan T. Padding, Selene Pirola, Willian Hogendoorn (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-25523.txt.
Introduction
Amiri and co-authors sharpen a bedside estimate of pressure loss across a narrowed artery. Clinics infer the drop from blood speed through Bernoulli formulas, but plain forms ignore how losses shift with flow regime. Their modified version adds a loss factor that moves with the Reynolds number. Bench tests on a model stenosis join direct pressure reads with ultrasound and MRI speed maps. The new form tracks the true drops best. Simpler forms skew high by wide margins. MRI pixel size also matters: coarse maps understate flow.
Catalog record
| CREATOR | Ali Amiri, Johan T. Padding, Selene Pirola, Willian Hogendoorn |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52510 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.25523
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Trans-stenotic pressure gradient estimation using a modified Bernoulli equation Ali Amiri, Johan T. Padding, Selene Pirola, Willian Hogendoorn Abstract Accurate non-invasive estimation of trans-stenotic pressure gradients remains a challenge. In clinical practice, pressure gradients are often estimated from velocity measurements using Bernoulli-based formulas, but these simplified relations do not explicitly account for how pressure losses change with the flow regime. Here, we introduce a modified Bernoulli (MB) formulation that incorporates regime-dependent pressure losses through a Reynolds-number-dependent loss coefficient. Steady in-vitro experiments were performed in an idealized stenosis model over physiologically relevant flow rates ( 0.65 − 3.9 0.65-3.9 L/min), combining direct pressure measurements with ultrasound imaging velocimetry (UIV) and phase-contrast magnetic resonance imaging (PC-MRI) to measure velocities. The MB model was calibrated from the measured pressure drops and then evaluated against the simplified Bernoulli (SB) and extended Bernoulli (EB) formulations. Over the tested flow regime, MB agreed best with the measurements (typically within about ± 10 % \pm 10\% ). SB and EB showed larger biases, with errors of roughly 10 − 55 % 10-55\% (SB) and - 15 − 25 15-25 % \% (EB), and overestimated the pressure drop in the clinically relevant range. We additionally quantified the effect of PC-MRI in-plane pixel size on MRI-based pressure estimates. Coarse sampling of the stenosis throat (only 3 − 4 3-4 pixels across the throat radius) led to systematic underestimation of flow rate and bulk velocity (about - 34 34 to - 44 % 44\% ) and, consequently, of the MB-predicted pressure drop (about - 52 52 to - 62 % 62\% ). In contrast, the peak throat velocity was substantially less sensitive to pixel size, resulting in smaller estimation errors (about - 13 13 to - 18.7 % 18.7\% ) when used as input for the MB. Overall, the results demonstrate that accounting for flow-regime-dependent loss mechanisms enhances pressure drop estimation, and that sufficient sampling of the stenotic throat is crucial for MRI-based flow rate and pressure drop estimation. In addition, peak-velocity-based MB pressure drop estimations are less sensitive to pixel size. Introduction Over the past few decades, cardiovascular diseases have become one of the top three causes of death around the world, with arterial stenosis ranking among the most significant contributors to global cardiovascular disease [ 40 ] . Stenosis is a common disease, affecting up to 5 \% of heart valves [ 20 ] and 10 \% of blood vessels [ 25 ] . It leads to pathological pressure-drops, which increase the heart workload and cause cardiovascular complications and mortality [ 34 ] . Therefore, pressure gradients serve as crucial indicators of stenosis severity. One of the major challenges in the management of stenosis is the accurate assessment of lesion severity [ 32 ] . The hemodynamic significance could be assessed by measuring the pressure drop across the narrowed region. However, accurate determination of trans-stenotic pressure gradients represents a critical challenge in cardiovascular medicine, directly influencing therapeutic decisions for patients with arterial stenoses and valvular disease. While catheter-based measurements remain the clinical gold standard, their invasive nature and associated risks have motivated extensive research into non-invasive alternatives [ 16 , 47 ] . The convergence of advanced medical imaging, computational methods, and refined analytical methods has created new possibilities for pressure gradient assessment. However, significant challenges persist in translating these approaches to routine clinical practice. 4D-flow MRI (time-resolved 3D phase-contrast magnetic resonance imaging) has become a key technique for non-invasive quantification of cardiovascular hemodynamics, providing volumetric velocity fields that are increasingly used to characterize stenotic jets, secondary flow structures, and clinically relevant flow-derived biomarkers [ 23 , 39 , 2 ] . Markl et al. outline both the clinical potential and the practical limitations of 4D-flow MRI for cardiovascular assessment [ 23 ] . Multiple studies emphasize that quantitative interpretation is strongly limited by acquisition constraints, most notably pixel size [ 9 , 16 , 8 ] . Larger pixel sizes can systematically bias measured velocities through partial-volume effects (PVE). This bias is amplified by smoothing of sharp gradients, particularly in narrow stenotic throats and high-shear regions. Collectively, this literature motivates explicitly accounting for MRI in-plane pixel size when using phase-contrast (PC)-MRI data as input to pressure drop estimation, because pressure losses scale strongly with characteristic velocities. Simplified formulations of the Bernoulli equation have long dominated clinical practice for non-invasive pressure-drop estimation, such as echocardiographic evaluation of valvular stenosis [ 3 ] . The simplified Bernoulli equation (SB) remains widely used, even though its limitations are well documented [ 13 , 19 , 16 , 31 , 35 , 28 ] . Kazemi et al. [ 19 ] evaluated relative pressure estimation in a stenotic phantom using 4D-flow MRI. They found that the pressure drop estimation from SB could deviate by 22 \% at moderate flow and 40 \% at the highest flow rate tested. Garcia et al. [ 12 ] advanced this model by introducing an extended Bernoulli equation (EB). In this model the ratio between the effective orifice area and the vessel area is incorporated. This effectively accounts for the downstream pressure recovery due to the jet deceleration. However, despite this improvement, the EB does not explicitly capture the change in flow physics from laminar, viscous-dominated losses to turbulent, inertia-dominated losses [ 44 ] . Turbulent losses and pressure recovery are not fully accounted for in the SB and EB models. Quantification of turbulence has emerged as a promising avenue to improve the estimation of trans-stenotic pressure gradient. Ha et al. [ 16 ] demonstrated that turbulence production defined using 4D-flow MRI correlates strongly with irreversible pressure-losses across stenoses. Similarly, Casas et al. [ 6 ] showed that quantification of turbulent kinetic energy (TKE) can provide geometry-dependent estimates of net trans-stenotic pressure gradients, with more severe stenoses exhibiting higher turbulent energy dissipation. This finding further supports the necessity to account for turbulent energy dissipation for an accurate estimation of lesion severity [ 10 ] , especially during exercise or pharmacologic stress. This link to turbulent energy dissipation highlights that the pressure-loss coefficient depends on the flow regime, which is typically characterized by the Reynolds number ( Re ). However, many clinically established estimation methods still assume a constant flow regime with independent correlations. This flow regime sensitivity has profound clinical implications, particularly for quantifying stenosis severity. To illustrate: a stenosis assessed as moderate at rest may behave quite differently during exercise when flow rates increase and the flow becomes turbulent [ 22 , 29 , 14 ] . Several studies have used analytical solutions for flow through constricted geometries to incorporate flow regime effects into the model [ 46 , 26 , 45 , 30 ] . Young and Tsai [ 46 ] indicated that pressure-loss mechanisms vary significantly across laminar, transitional, and turbulent regimes. For steady flow they showed that the pressure gradient can be written as the sum of a viscous term and a turbulence dissipation term. This approach was later extended to account for the effects of pulsatile flow [ 45 ] and stenosis shape [ 37 ] . However, their correlation, expressed solely in terms of Re , cannot be directly applied in routine clinical settings particularly with Doppler echocardiography and PC-MRI,