Velocity encoding versus acceleration encoding for pressure gradient estimation in MR haemodynamic studies

Area of Science:

• Medical Imaging
• Fluid Dynamics
• Biomedical Engineering

Background:

• Extracting pressure gradient maps from magnetic resonance (MR) images is crucial for understanding hemodynamics.
• Existing methods often rely on the Navier-Stokes model and require knowledge of the flow acceleration.
• Traditional approaches calculate acceleration from spatial and temporal derivatives of MR velocity data.

Purpose of the Study:

• To compare the accuracy of directly measured MR acceleration fields with those derived from MR velocity acquisitions.
• To evaluate the inertial and convective components of acceleration using experimental phantoms.
• To investigate the sources of noise and artifacts in MR-based acceleration measurements.

Main Methods:

• Direct MR measurement of acceleration field components.
• Calculation of acceleration from temporal and spatial derivatives of MR velocity data.
• Utilized two experimental phantoms (inertial and convective) and mathematical simulation.

Main Results:

• Direct MR measurement of acceleration fields resulted in lower noise levels.
• Direct measurement produced fewer artifacts compared to calculations from velocity derivatives.
• Mathematical simulations elucidated noise origins from velocity derivatives and MR sequence misregistration.

Conclusions:

• Direct MR measurement of acceleration is a more robust method for pressure gradient map extraction.
• This technique offers improved accuracy by minimizing noise and artifacts inherent in velocity-derived methods.
• The findings support the adoption of direct acceleration measurement in hemodynamic MR studies.