% 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{Hekler:144770,
      author       = {A. Hekler$^*$ and J. S. Utikal$^*$ and A. H. Enk and A.
                      Hauschild and M. Weichenthal and R. C. Maron and C. Berking
                      and S. Haferkamp and J. Klode and D. Schadendorf and B.
                      Schilling and T. Holland-Letz$^*$ and B. Izar and C. von
                      Kalle$^*$ and S. Fröhling$^*$ and T. J. Brinker$^*$ and L.
                      Schmitt and W. K. Peitsch and F. Hoffmann and J. C. Becker
                      and C. Drusio and P. Jansen and J. Klode and G. Lodde and S.
                      Sammet and D. Schadendorf and W. Sondermann and S. Ugurel
                      and J. Zader and A. Enk and M. Salzmann and S. Schäfer and
                      K. Schäkel and J. Winkler and P. Wölbing and H. Asper and
                      A.-S. Bohne and V. Brown and B. Burba and S. Deffaa and C.
                      Dietrich and M. Dietrich and K. A. Drerup and F. Egberts and
                      A.-S. Erkens and S. Greven and V. Harde and M. Jost and M.
                      Kaeding and K. Kosova and S. Lischner and M. Maagk and A. L.
                      Messinger and M. Metzner and R. Motamedi and A.-C. Rosenthal
                      and U. Seidl and J. Stemmermann and K. Torz and J. G. Velez
                      and J. Haiduk and M. Alter and C. Bär and P. Bergenthal and
                      A. Gerlach and C. Holtorf and A. Karoglan and S. Kindermann
                      and L. Kraas and M. Felcht and M. R. Gaiser and C.-D. Klemke
                      and H. Kurzen and T. Leibing and V. Müller and R. R.
                      Reinhard and J. Utikal$^*$ and F. Winter and C. Berking and
                      L. Eicher and D. Hartmann and M. Heppt and K. Kilian and S.
                      Krammer and D. Lill and A.-C. Niesert and E. Oppel and E.
                      Sattler and S. Senner and J. Wallmichrath and H. Wolff and
                      A. Gesierich and T. Giner and V. Glutsch and A. Kerstan and
                      D. Presser and P. Schrüfer and P. Schummer and I. Stolze
                      and J. Weber and K. Drexler and S. Haferkamp and M. Mickler
                      and C. T. Stauner and A. Thiem},
      collaboration = {Collaborators},
      title        = {{S}uperior skin cancer classification by the combination of
                      human and artificial intelligence.},
      journal      = {European journal of cancer},
      volume       = {120},
      issn         = {0959-8049},
      address      = {Amsterdam [u.a.]},
      publisher    = {Elsevier},
      reportid     = {DKFZ-2019-02202},
      pages        = {114 - 121},
      year         = {2019},
      abstract     = {In recent studies, convolutional neural networks (CNNs)
                      outperformed dermatologists in distinguishing dermoscopic
                      images of melanoma and nevi. In these studies,
                      dermatologists and artificial intelligence were considered
                      as opponents. However, the combination of classifiers
                      frequently yields superior results, both in machine learning
                      and among humans. In this study, we investigated the
                      potential benefit of combining human and artificial
                      intelligence for skin cancer classification.Using 11,444
                      dermoscopic images, which were divided into five diagnostic
                      categories, novel deep learning techniques were used to
                      train a single CNN. Then, both 112 dermatologists of 13
                      German university hospitals and the trained CNN
                      independently classified a set of 300 biopsy-verified skin
                      lesions into those five classes. Taking into account the
                      certainty of the decisions, the two independently determined
                      diagnoses were combined to a new classifier with the help of
                      a gradient boosting method. The primary end-point of the
                      study was the correct classification of the images into five
                      designated categories, whereas the secondary end-point was
                      the correct classification of lesions as either benign or
                      malignant (binary classification).Regarding the multiclass
                      task, the combination of man and machine achieved an
                      accuracy of $82.95\%.$ This was $1.36\%$ higher than the
                      best of the two individual classifiers $(81.59\%$ achieved
                      by the CNN). Owing to the class imbalance in the binary
                      problem, sensitivity, but not accuracy, was examined and
                      demonstrated to be superior $(89\%)$ to the best individual
                      classifier (CNN with $86.1\%).$ The specificity in the
                      combined classifier decreased from $89.2\%$ to $84\%.$
                      However, at an equal sensitivity of $89\%,$ the CNN achieved
                      a specificity of only $81.5\%$ INTERPRETATION: Our findings
                      indicate that the combination of human and artificial
                      intelligence achieves superior results over the independent
                      results of both of these systems.},
      cin          = {A370 / C060 / B340},
      ddc          = {610},
      cid          = {I:(DE-He78)A370-20160331 / I:(DE-He78)C060-20160331 /
                      I:(DE-He78)B340-20160331},
      pnm          = {311 - Signalling pathways, cell and tumor biology
                      (POF3-311)},
      pid          = {G:(DE-HGF)POF3-311},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:31518967},
      doi          = {10.1016/j.ejca.2019.07.019},
      url          = {https://inrepo02.dkfz.de/record/144770},
}