What is it about?
We use neural networks informed with the RANS equations and boundary conditions to correct an artifact in 4D flow MRI data. By adding equations and unknowns, the networks is able to assimilate new variables, such as the mean pressure and Reynolds stresses.
Featured Image
Photo by Cassi Josh on Unsplash
Why is it important?
This study demonstrates the effectiveness of PINNs for artifact correction and data assimilation in a simplified steady flow in a rigid axisymmetric stenosis.
Perspectives
We want to go toward realistic clinical applications and extend the method to 3D pulsatile flow in arbitrary geometries.
Alexandre Villié
Technische Universitat Berlin
Read the Original
This page is a summary of: Physics-informed neural networks for enhancing medical flow magnetic resonance imaging: Artifact correction and mean pressure and Reynolds stresses assimilation, Physics of Fluids, February 2025, American Institute of Physics,
DOI: 10.1063/5.0252852.
You can read the full text:
Contributors
The following have contributed to this page
