% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@ARTICLE{MaierHein:288083,
      author       = {L. Maier-Hein$^*$ and A. Reinke$^*$ and P. Godau$^*$ and M.
                      D. Tizabi$^*$ and F. Büttner$^*$ and E. Christodoulou$^*$
                      and B. Glocker and F. Isensee$^*$ and J. Kleesiek and M.
                      Kozubek and M. Reyes and M. A. Riegler and M.
                      Wiesenfarth$^*$ and A. E. Kavur$^*$ and C. H. Sudre and M.
                      Baumgartner$^*$ and M. Eisenmann$^*$ and D.
                      Heckmann-Nötzel$^*$ and T. Rädsch$^*$ and L. Acion and M.
                      Antonelli and T. Arbel and S. Bakas and A. Benis and M. B.
                      Blaschko and M. J. Cardoso and V. Cheplygina and B. A.
                      Cimini and G. S. Collins and K. Farahani and L. Ferrer and
                      A. Galdran and B. van Ginneken and R. Haase and D. A.
                      Hashimoto and M. M. Hoffman and M. Huisman and P. Jannin and
                      C. E. Kahn and D. Kainmueller and B. Kainz and A. Karargyris
                      and A. Karthikesalingam and F. Kofler and A.
                      Kopp-Schneider$^*$ and A. Kreshuk and T. Kurc and B. A.
                      Landman and G. Litjens and A. Madani and K. Maier-Hein$^*$
                      and A. L. Martel and P. Mattson and E. Meijering and B.
                      Menze and K. G. M. Moons and H. Müller and B. Nichyporuk
                      and F. Nickel and J. Petersen and N. Rajpoot and N. Rieke
                      and J. Saez-Rodriguez and C. I. Sánchez and S. Shetty and
                      M. van Smeden and R. M. Summers and A. A. Taha and A.
                      Tiulpin and S. A. Tsaftaris and B. Van Calster and G.
                      Varoquaux and P. Jäger$^*$},
      title        = {{M}etrics reloaded: recommendations for image analysis
                      validation.},
      journal      = {Nature methods},
      volume       = {21},
      number       = {2},
      issn         = {1548-7091},
      address      = {London [u.a.]},
      publisher    = {Nature Publishing Group},
      reportid     = {DKFZ-2024-00337},
      pages        = {195 - 212},
      year         = {2024},
      note         = {#EA:E130#LA:E290#},
      abstract     = {Increasing evidence shows that flaws in machine learning
                      (ML) algorithm validation are an underestimated global
                      problem. In biomedical image analysis, chosen performance
                      metrics often do not reflect the domain interest, and thus
                      fail to adequately measure scientific progress and hinder
                      translation of ML techniques into practice. To overcome
                      this, we created Metrics Reloaded, a comprehensive framework
                      guiding researchers in the problem-aware selection of
                      metrics. Developed by a large international consortium in a
                      multistage Delphi process, it is based on the novel concept
                      of a problem fingerprint-a structured representation of the
                      given problem that captures all aspects that are relevant
                      for metric selection, from the domain interest to the
                      properties of the target structure(s), dataset and algorithm
                      output. On the basis of the problem fingerprint, users are
                      guided through the process of choosing and applying
                      appropriate validation metrics while being made aware of
                      potential pitfalls. Metrics Reloaded targets image analysis
                      problems that can be interpreted as classification tasks at
                      image, object or pixel level, namely image-level
                      classification, object detection, semantic segmentation and
                      instance segmentation tasks. To improve the user experience,
                      we implemented the framework in the Metrics Reloaded online
                      tool. Following the convergence of ML methodology across
                      application domains, Metrics Reloaded fosters the
                      convergence of validation methodology. Its applicability is
                      demonstrated for various biomedical use cases.},
      subtyp        = {Review Article},
      cin          = {E130 / FM01 / E230 / C060 / E290},
      ddc          = {610},
      cid          = {I:(DE-He78)E130-20160331 / I:(DE-He78)FM01-20160331 /
                      I:(DE-He78)E230-20160331 / I:(DE-He78)C060-20160331 /
                      I:(DE-He78)E290-20160331},
      pnm          = {315 - Bildgebung und Radioonkologie (POF4-315)},
      pid          = {G:(DE-HGF)POF4-315},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:38347141},
      doi          = {10.1038/s41592-023-02151-z},
      url          = {https://inrepo02.dkfz.de/record/288083},
}