%0 Journal Article
%A Brinker, Titus J
%A Hekler, Achim
%A Enk, Alexander H
%A Klode, Joachim
%A Hauschild, Axel
%A Berking, Carola
%A Schilling, Bastian
%A Haferkamp, Sebastian
%A Schadendorf, Dirk
%A Holland-Letz, Tim
%A Utikal, Jochen S
%A von Kalle, Christof
%A Ludwig-Peitsch, Wiebke
%A Sirokay, Judith
%A Heinzerling, Lucie
%A Albrecht, Magarete
%A Baratella, Katharina
%A Bischof, Lena
%A Chorti, Eleftheria
%A Dith, Anna
%A Drusio, Christina
%A Giese, Nina
%A Gratsias, Emmanouil
%A Griewank, Klaus
%A Hallasch, Sandra
%A Hanhart, Zdenka
%A Herz, Saskia
%A Hohaus, Katja
%A Jansen, Philipp
%A Jockenhöfer, Finja
%A Kanaki, Theodora
%A Knispel, Sarah
%A Leonhard, Katja
%A Martaki, Anna
%A Matei, Liliana
%A Matull, Johanna
%A Olischewski, Alexandra
%A Petri, Maximilian
%A Placke, Jan-Malte
%A Raub, Simon
%A Salva, Katrin
%A Schlott, Swantje
%A Sody, Elsa
%A Steingrube, Nadine
%A Stoffels, Ingo
%A Ugurel, Selma
%A Zaremba, Anne
%A Gebhardt, Christoffer
%A Booken, Nina
%A Christolouka, Maria
%A Buder-Bakhaya, Kristina
%A Bokor-Billmann, Therezia
%A Enk, Alexander
%A Gholam, Patrick
%A Hänßle, Holger
%A Salzmann, Martin
%A Schäfer, Sarah
%A Schäkel, Knut
%A Schank, Timo
%A Bohne, Ann-Sophie
%A Deffaa, Sophia
%A Drerup, Katharina
%A Egberts, Friederike
%A Erkens, Anna-Sophie
%A Ewald, Benjamin
%A Falkvoll, Sandra
%A Gerdes, Sascha
%A Harde, Viola
%A Hauschild, Axel
%A Jost, Marion
%A Kosova, Katja
%A Messinger, Laetitia
%A Metzner, Malte
%A Morrison, Kirsten
%A Motamedi, Rogina
%A Pinczker, Anja
%A Rosenthal, Anne
%A Scheller, Natalie
%A Schwarz, Thomas
%A Stölzl, Dora
%A Thielking, Federieke
%A Tomaschewski, Elena
%A Wehkamp, Ulrike
%A Weichenthal, Michael
%A Wiedow, Oliver
%A Bär, Claudia Maria
%A Bender-Säbelkampf, Sophia
%A Horbrügger, Marc
%A Karoglan, Ante
%A Kraas, Luise
%A Faulhaber, Jörg
%A Geraud, Cyrill
%A Guo, Ze
%A Koch, Philipp
%A Linke, Miriam
%A Maurier, Nolwenn
%A Müller, Verena
%A Thomas, Benjamin
%A Utikal, Jochen Sven
%A Alamri, Ali Saeed M
%A Baczako, Andrea
%A Berking, Carola
%A Betke, Matthias
%A Haas, Carolin
%A Hartmann, Daniela
%A Heppt, Markus V
%A Kilian, Katharina
%A Krammer, Sebastian
%A Lapczynski, Natalie Lidia
%A Mastnik, Sebastian
%A Nasifoglu, Suzan
%A Ruini, Cristel
%A Sattler, Elke
%A Schlaak, Max
%A Wolff, Hans
%A Achatz, Birgit
%A Bergbreiter, Astrid
%A Drexler, Konstantin
%A Ettinger, Monika
%A Haferkamp, Sebastian
%A Halupczok, Anna
%A Hegemann, Marie
%A Dinauer, Verena
%A Maagk, Maria
%A Mickler, Marion
%A Philipp, Biance
%A Wilm, Anna
%A Wittmann, Constanze
%A Gesierich, Anja
%A Glutsch, Valerie
%A Kahlert, Katrin
%A Kerstan, Andreas
%A Schilling, Bastian
%A Schrüfer, Philipp
%T Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task.
%J European journal of cancer
%V 113
%@ 0959-8049
%C Amsterdam [u.a.]
%I Elsevier
%M DKFZ-2019-01167
%P 47 - 54
%D 2019
%X Recent studies have successfully demonstrated the use of deep-learning algorithms for dermatologist-level classification of suspicious lesions by the use of excessive proprietary image databases and limited numbers of dermatologists. For the first time, the performance of a deep-learning algorithm trained by open-source images exclusively is compared to a large number of dermatologists covering all levels within the clinical hierarchy.We used methods from enhanced deep learning to train a convolutional neural network (CNN) with 12,378 open-source dermoscopic images. We used 100 images to compare the performance of the CNN to that of the 157 dermatologists from 12 university hospitals in Germany. Outperformance of dermatologists by the deep neural network was measured in terms of sensitivity, specificity and receiver operating characteristics.The mean sensitivity and specificity achieved by the dermatologists with dermoscopic images was 74.1
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:30981091
%R 10.1016/j.ejca.2019.04.001
%U https://inrepo02.dkfz.de/record/143587