We report the design, protocol, and outcomes of a student reproducibility hackathon focused on replicating the results of three influential MRI reconstruction papers: (a) MoDL, an unrolled model-based network with learned denoising; (b) HUMUS-Net, a hybrid unrolled multiscale CNN+Transformer architecture; and (c) an untrained, physics-regularized dynamic MRI method that uses a quantitative MR model for early stopping. We describe the setup of the hackathon and present reproduction outcomes alongside additional experiments, and we detail fundamental practices for building reproducible codebases.
@article{arxiv.2601.18314,
title = {A Master Class on Reproducibility: A Student Hackathon on Advanced MRI Reconstruction Methods},
author = {Lina Felsner and Sevgi G. Kafali and Hannah Eichhorn and Agnes A. J. Leth and Aidas Batvinskas and Andre Datchev and Fabian Klemm and Jan Aulich and Puntika Leepagorn and Ruben Klinger and Daniel Rueckert and Julia A. Schnabel},
journal= {arXiv preprint arXiv:2601.18314},
year = {2026}
}