Simulation-based parameter optimization for fetal brain MRI super-resolution reconstruction

TL;DR

This paper introduces a simulation-based method to optimize the regularization parameter in fetal brain MRI super-resolution, improving reconstruction accuracy by tailoring it to specific acquisition settings.

Contribution

It proposes a novel simulation-based approach to tune the hyperparameter, addressing the challenge of lack of ground truth in fetal MRI super-resolution.

Findings

Optimized regularization parameter improves reconstruction accuracy.

Simulation-based tuning outperforms default values.

Qualitative validation confirms clinical relevance.

Abstract

Tuning the regularization hyperparameter $\alpha$ in inverse problems has been a longstanding problem. This is particularly true in the case of fetal brain magnetic resonance imaging, where an isotropic high-resolution volume is reconstructed from motion-corrupted low-resolution series of two-dimensional thick slices. Indeed, the lack of ground truth images makes challenging the adaptation of $\alpha$ to a given setting of interest in a quantitative manner. In this work, we propose a simulation-based approach to tune $\alpha$ for a given acquisition setting. We focus on the influence of the magnetic field strength and availability of input low-resolution images on the ill-posedness of the problem. Our results show that the optimal $\alpha$, chosen as the one maximizing the similarity with the simulated reference image, significantly improves the super-resolution reconstruction accuracy compared to the generally adopted default regularization values, independently of the selected pipeline. Qualitative validation on clinical data confirms the importance of tuning this parameter to the targeted clinical image setting.