Hippocampal subfield segmentation requires high-resolution T2w turbo spin echo (TSE) MRI, yet this sequence is susceptible to motion artifacts, leading to substantial data loss. We developed a conditional generative model (MRecover) that synthesizes routinely acquired T1w images to create TSE images with autoregressive slice conditioning for volumetric consistency. Trained on 7T MRI data (n=577), the model achieved high in-domain fidelity (n=148, SSIM=0.84, FSIM=0.94) and generalized well to out-of-domain 3T data: subfield volumes from synthesized and the as-acquired images closely matched: (n=416, r=0.87-0.97) and yielded 31.8% more analyzable subjects in the motion-affected ADNI3 dataset after quality control (593 vs 450). The synthesized images also achieved larger effect sizes due to increasing the sample size for diagnostic group differences in hippocampal subfield atrophy (whole hippocampus ϵ2= 0.121-0.100 vs. 0.086-0.062, left-right hemispheres). Project page: https://jinghangli98.github.io/MRecover/
@article{arxiv.2605.21669,
title = {MRecover: A Conditional Generative Model for Recovering Motion-Corrupted MR images Using AI Generated Contrast},
author = {Jinghang Li and Tales Santini and Courtney Clark and Bruno de Almeida and Cong Chu and Salem Alkhateeb and Andrea Sajewski and Jacob Berardinelli and Hecheng Jin and Tobias Campos and Jeremy J. Berardo and Joseph Mettenburg and Ariel Gildengers and Howard J. Aizenstein and Minjie Wu and Tamer S. Ibrahim},
journal= {arXiv preprint arXiv:2605.21669},
year = {2026}
}