Motivation: 4D flow MRI suffers from different sources of noise and wrapping artifacts which can affect its accuracy and usability as a clinical tool. Goal(s): This study aims to simultaneously improve signal-to-noise ratio and fix velocity wrapping artifacts in 4D Flow MRI. Approach: We developed an unsupervised neural network that enhances 4D Flow MRI by estimating a divergence-free velocity field. Results: The model demonstrated superior performance compared to existing methods, and initial in vivo results validated its potential for more reliable, artifact-free hemodynamic assessments in clinical applications. Impact: We proposed an unsupervised divergence-free neural network that effectively enhances the signal-to-noise ratio and reduces velocity wrapping artifacts in 4D Flow MRI, improving its accuracy and reliability in both clinical and research settings
Bisbal et al. (Tue,) studied this question.