Scholarlink

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 [1]; (b) HUMUS-Net, a hybrid unrolled multi-scale CNN+Transformer architecture [2]; and (c) an untrained, physics-regularized dynamic MRI method that uses a quantitativeMRmodel for early stopping [3]. We describe the setup of the hackathon and present reproduction outcomes alongside additional experiments, and we detail fundamental practices for building reproducible codebases.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Master Class on Reproducibility

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Lina Felsner,
  • Sevgi G. Kafali,
  • Hannah Eichhorn,
  • Agnes A. J. Leth,
  • Aidas Batvinskas,
  • André Datchev,
  • Fabian Klemm,
  • Jan Aulich,
  • Puntika Leepagorn,
  • Ruben Klinger,
  • Daniel Rueckert,
  • Julia A. Schnabel

摘要

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 [1]; (b) HUMUS-Net, a hybrid unrolled multi-scale CNN+Transformer architecture [2]; and (c) an untrained, physics-regularized dynamic MRI method that uses a quantitativeMRmodel for early stopping [3]. We describe the setup of the hackathon and present reproduction outcomes alongside additional experiments, and we detail fundamental practices for building reproducible codebases.