% 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{Chanda:301729,
      author       = {T. Chanda$^*$ and S. Haggenmueller$^*$ and T. Bucher$^*$
                      and T. Holland-Letz$^*$ and H. Kittler and P. Tschandl and
                      M. V. Heppt and C. Berking and J. Utikal$^*$ and B.
                      Schilling and C. Buerger and C. Navarrete-Dechent and M.
                      Goebeler and J. N. Kather and C. V. Schneider and B. Durani
                      and H. Durani and M. Jansen and J. Wacker and J. Wacker and
                      T. Brinker$^*$},
      collaboration = {R. S. Consortium},
      othercontributors = {N. Booken and V. Ahlgrimm-Siess and J. Welzel and O.-D.
                          Persa and F. Dimitriou and S. A. Braun and L. V. Maul and A.
                          Reimer-Taschenbrecker and S. Schuh and F. G. Bechara and L.
                          Feldmeyer and B. Mühleisen and E. Gössinger and S. A.
                          Braun and V. A. Nguyen and J.-T. Maul and F. Hoffmann and C.
                          Pföhler and J. Thamm and W. Ludwig-Peitsch and D. Hartmann
                          and L. Garzona-Navas and M. Sławińska and P.
                          Theofilogiannakou and A. S. Vucemilovic and J. J.
                          Lluch-Galcerá and A. Beyens and D. I. Erdil and R. Afiouni
                          and V. Bondare-Ansberga and M. A. Morales-Sánchez and A.
                          Ferhatosmanoğlu and R. R. O. Neto and L. Petrovska and A.
                          Tsakiri and H. Cenk and S. Hudson and M. Dragolov and Z.
                          Zafirovik and I. Jocic and A. Balcere and Z. Lengyel and A.
                          Salava and I. Hoorens and S. R. Saa and E. Rácz and G.
                          Salerni and K. Manuelyan and A. M. Ammar and M. Erdmann and
                          N. Wagner and J. Sambale and S. Kemenes and M. Ronicke and
                          L. Sollfrank and C. Bosch-Voskens and I. Sagonas and T.
                          Breakell and C. Uebel and L. Zieringer and M. Hoener and L.
                          Rabe and T. Sackmann and J. Baumert and M. L. Schaarschmidt
                          and N. Ninosu and K. Yilmaz and D. Dionysia and F. Christ
                          and S. Fahimi and S. Loos and A. Sachweizer and J. Gosmann
                          and T. Weberschock and U. Erdogdu and A. Buchinger and J.
                          Lunderstedt and T. Funk and H. Klifo and S. Kiefer and D.
                          Klifo and M. Kalski},
      title        = {{D}ermatologist-like explainable {AI} enhances melanoma
                      diagnosis accuracy: eye-tracking study.},
      journal      = {Nature Communications},
      volume       = {16},
      number       = {1},
      issn         = {2041-1723},
      address      = {[London]},
      publisher    = {Springer Nature},
      reportid     = {DKFZ-2025-01121},
      pages        = {4739},
      year         = {2025},
      note         = {#EA:C140#LA:C140#},
      abstract     = {Artificial intelligence (AI) systems substantially improve
                      dermatologists' diagnostic accuracy for melanoma, with
                      explainable AI (XAI) systems further enhancing their
                      confidence and trust in AI-driven decisions. Despite these
                      advancements, there remains a critical need for objective
                      evaluation of how dermatologists engage with both AI and XAI
                      tools. In this study, 76 dermatologists participate in a
                      reader study, diagnosing 16 dermoscopic images of melanomas
                      and nevi using an XAI system that provides detailed,
                      domain-specific explanations, while eye-tracking technology
                      assesses their interactions. Diagnostic performance is
                      compared with that of a standard AI system lacking
                      explanatory features. Here we show that XAI significantly
                      improves dermatologists' diagnostic balanced accuracy by 2.8
                      percentage points compared to standard AI. Moreover,
                      diagnostic disagreements with AI/XAI systems and complex
                      lesions are associated with elevated cognitive load, as
                      evidenced by increased ocular fixations. These insights have
                      significant implications for the design of AI/XAI tools for
                      visual tasks in dermatology and the broader development of
                      XAI in medical diagnostics.},
      keywords     = {Melanoma: diagnosis / Melanoma: diagnostic imaging / Humans
                      / Artificial Intelligence / Eye-Tracking Technology /
                      Dermatologists / Skin Neoplasms: diagnosis / Skin Neoplasms:
                      diagnostic imaging / Female / Dermoscopy: methods / Male /
                      Nevus: diagnosis / Nevus: diagnostic imaging / Adult /
                      Middle Aged},
      cin          = {C140 / C060 / A370},
      ddc          = {500},
      cid          = {I:(DE-He78)C140-20160331 / I:(DE-He78)C060-20160331 /
                      I:(DE-He78)A370-20160331},
      pnm          = {313 - Krebsrisikofaktoren und Prävention (POF4-313)},
      pid          = {G:(DE-HGF)POF4-313},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:40399272},
      pmc          = {pmc:PMC12095463},
      doi          = {10.1038/s41467-025-59532-5},
      url          = {https://inrepo02.dkfz.de/record/301729},
}