A Master Class on Reproducibility: A Student Hackathon on Advanced MRI Reconstruction Methods
MCML Authors
Lina Felsner
Dr.
Daniel Rückert
Prof. Dr.
Director
Julia Schnabel
Prof. Dr.
Principal Investigator
Abstract
Lina Felsner
Dr.
Daniel Rückert
Prof. Dr.
Director
Julia Schnabel
Prof. Dr.
Principal Investigator
Abstract
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.
misc FKE+26
Preprint
Jan. 2026Authors
L. Felsner • S. G. Kafali • H. Eichhorn • A. A. J. Leth • A. Batvinskas • A. Datchev • F. Klemm • J. Aulich • P. Leepagorn • R. Klinger • D. Rückert • J. A. SchnabelLinks
arXivResearch Area
BibTeXKey: FKE+26