[Paper Review] 4D-Flow MRI Pressure Estimation Using Velocity Measurement-Error based Weighted Least-Squares

Jiacheng Zhang, Melissa C. Brindise+6 more|arXiv (Cornell University)|Apr 30, 2019

Advanced MRI Techniques and ApplicationsMedicine20 references3 citations

TL;DR

This paper proposes a velocity measurement-error based weighted least-squares (WLS) method for 4D-Flow MRI pressure reconstruction, improving accuracy by estimating velocity errors from velocity divergence and propagating them through the Navier-Stokes equations to generate error-weighted pressure integration. The method reduces pressure error by 50% to over 200% in spatially varying error conditions and shows superior agreement with in vitro and in vivo data compared to the traditional pressure Poisson equation approach.

ABSTRACT

This work introduces a 4D-flow magnetic resonance imaging (MRI) pressure reconstruction method which employs weighted least-squares (WLS) for pressure integration. Pressure gradients are calculated from the velocity fields, and velocity errors are estimated from the velocity divergence for incompressible flow. Pressure gradient errors are estimated by propagating the velocity errors through Navier-Stokes momentum equation. A weight matrix is generated based on the pressure gradient errors, then employed for pressure reconstruction. The pressure reconstruction method was demonstrated and analyzed using synthetic velocity fields as well as Poiseuille flow measured using in vitro 4D-flow MRI. Performance of the proposed WLS method was compared to the method of solving the pressure Poisson equation which has been the primary method used in the previous studies. Error analysis indicated that the proposed method is more robust to velocity measurement errors. Improvement on pressure results was found to be more significant for the cases with spatially-varying velocity error level, with reductions in error ranging from 50% to over 200%. Finally, the method was applied to flow in a patient-specific cerebral aneurysm. Validation was performed with in vitro flow data collected using Particle Tracking Velocimetry (PTV) and Shake the Box (STB) method, and in vivo flow measurement obtained using 4D-flow MRI. Pressure calculated by WLS, as opposed to the Poisson equation, was more consistent with the flow structures and showed better agreement between the in vivo and in vitro data. These results suggest the utility of WLS method to obtain reliable pressure field from clinical flow measurement data.

Motivation & Objective

• To address the challenge of inaccurate pressure reconstruction in 4D-Flow MRI due to velocity measurement errors.
• To improve robustness of pressure field estimation in hemodynamically complex flows, such as in cerebral aneurysms.
• To develop a method that accounts for spatially varying velocity error distributions in clinical 4D-Flow MRI data.
• To reduce pressure reconstruction error by incorporating error-weighted least-squares integration based on estimated velocity gradient uncertainties.
• To validate the method against in vitro PTV and STB data and in vivo 4D-Flow MRI, demonstrating improved consistency with flow structures.

Proposed method

• Velocity errors are estimated from the divergence of the velocity field under the assumption of incompressible flow.
• Pressure gradient errors are computed by propagating velocity errors through the Navier-Stokes momentum equations.
• A weight matrix is constructed from the estimated pressure gradient errors to reflect measurement uncertainty in the integration process.
• Weighted least-squares (WLS) is applied to integrate pressure gradients, minimizing the impact of high-error regions.
• The method avoids solving the pressure Poisson equation, which is sensitive to velocity noise and spatially non-uniform errors.
• The approach is validated using synthetic velocity fields, in vitro Poiseuille flow data, and in vivo patient-specific cerebral aneurysm data.

Experimental results

Research questions

• RQ1 How does the proposed WLS method with error-weighted integration improve pressure reconstruction accuracy compared to the pressure Poisson equation in the presence of velocity measurement errors?
• RQ2 To what extent does the method reduce pressure error when velocity errors vary spatially across the imaging domain?
• RQ3 How well does the WLS-based pressure estimation agree with in vitro flow measurements obtained via PTV and STB in a patient-specific aneurysm model?
• RQ4 Does the WLS method produce pressure fields that are more consistent with observed hemodynamic structures than the Poisson-based method?
• RQ5 Can the WLS method reliably reconstruct pressure fields from clinical 4D-Flow MRI data with realistic noise and resolution limitations?

Key findings

• The proposed WLS method reduced pressure reconstruction error by 50% to over 200% compared to the Poisson equation in cases with spatially varying velocity error levels.
• The method demonstrated significantly improved robustness to velocity measurement errors, particularly in regions with high local noise or inhomogeneous resolution.
• In the in vitro patient-specific aneurysm model, WLS-derived pressure fields showed better agreement with PTV and STB measurements than Poisson-based results.
• The WLS method produced pressure fields that were more consistent with complex flow structures, such as vortices and flow separation zones, than the Poisson method.
• The error propagation framework based on velocity divergence and Navier-Stokes equations provided a reliable estimate of pressure gradient uncertainty for weighting.
• The method outperformed the Poisson equation in both synthetic and real-world flow scenarios, confirming its utility for clinical 4D-Flow MRI applications.