Achim Hekler1, Jochen S Utikal2, Alexander H Enk3, Wiebke Solass4, Max Schmitt1, Joachim Klode5, Dirk Schadendorf5, Wiebke Sondermann5, Cindy Franklin6, Felix Bestvater7, Michael J Flaig8, Dieter Krahl9, Christof von Kalle1, Stefan Fröhling1, Titus J Brinker10. 1. National Center for Tumor Diseases, German Cancer Research Center, Heidelberg, Germany. 2. Department of Dermatology, Heidelberg University, Mannheim, Germany; Skin Cancer Unit, German Cancer Research Center, Heidelberg, Germany. 3. Department of Dermatology, University Hospital Heidelberg, Heidelberg, Germany. 4. Institute of Pathology and Neuropathology, Eberhard-Karls-University Tuebingen and National Center for Pleura and Peritoneum, University of Tuebingen, Germany. 5. Department of Dermatology, University Hospital Essen, Essen, Germany. 6. Department of Dermatology, University Hospital Cologne, Cologne, Germany. 7. Core Facility Unit Light Microscopy, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany. 8. Department of Dermatology, University Hospital Munich (LMU), Munich, Germany. 9. Private Laboratory of Dermatohistopathology, Mönchhofstraße 52, 69120 Heidelberg. 10. National Center for Tumor Diseases, German Cancer Research Center, Heidelberg, Germany; Department of Dermatology, University Hospital Heidelberg, Heidelberg, Germany. Electronic address: titus.brinker@dkfz.de.
Abstract
BACKGROUND: The diagnosis of most cancers is made by a board-certified pathologist based on a tissue biopsy under the microscope. Recent research reveals a high discordance between individual pathologists. For melanoma, the literature reports on 25-26% of discordance for classifying a benign nevus versus malignant melanoma. A recent study indicated the potential of deep learning to lower these discordances. However, the performance of deep learning in classifying histopathologic melanoma images was never compared directly to human experts. The aim of this study is to perform such a first direct comparison. METHODS: A total of 695 lesions were classified by an expert histopathologist in accordance with current guidelines (350 nevi/345 melanoma). Only the haematoxylin & eosin (H&E) slides of these lesions were digitalised via a slide scanner and then randomly cropped. A total of 595 of the resulting images were used to train a convolutional neural network (CNN). The additional 100 H&E image sections were used to test the results of the CNN in comparison to 11 histopathologists. Three combined McNemar tests comparing the results of the CNNs test runs in terms of sensitivity, specificity and accuracy were predefined to test for significance (p < 0.05). FINDINGS: The CNN achieved a mean sensitivity/specificity/accuracy of 76%/60%/68% over 11 test runs. In comparison, the 11 pathologists achieved a mean sensitivity/specificity/accuracy of 51.8%/66.5%/59.2%. Thus, the CNN was significantly (p = 0.016) superior in classifying the cropped images. INTERPRETATION: With limited image information available, a CNN was able to outperform 11 histopathologists in the classification of histopathological melanoma images and thus shows promise to assist human melanoma diagnoses.
BACKGROUND: The diagnosis of most cancers is made by a board-certified pathologist based on a tissue biopsy under the microscope. Recent research reveals a high discordance between individual pathologists. For melanoma, the literature reports on 25-26% of discordance for classifying a benign nevus versus malignant melanoma. A recent study indicated the potential of deep learning to lower these discordances. However, the performance of deep learning in classifying histopathologic melanoma images was never compared directly to human experts. The aim of this study is to perform such a first direct comparison. METHODS: A total of 695 lesions were classified by an expert histopathologist in accordance with current guidelines (350 nevi/345 melanoma). Only the haematoxylin & eosin (H&E) slides of these lesions were digitalised via a slide scanner and then randomly cropped. A total of 595 of the resulting images were used to train a convolutional neural network (CNN). The additional 100 H&E image sections were used to test the results of the CNN in comparison to 11 histopathologists. Three combined McNemar tests comparing the results of the CNNs test runs in terms of sensitivity, specificity and accuracy were predefined to test for significance (p < 0.05). FINDINGS: The CNN achieved a mean sensitivity/specificity/accuracy of 76%/60%/68% over 11 test runs. In comparison, the 11 pathologists achieved a mean sensitivity/specificity/accuracy of 51.8%/66.5%/59.2%. Thus, the CNN was significantly (p = 0.016) superior in classifying the cropped images. INTERPRETATION: With limited image information available, a CNN was able to outperform 11 histopathologists in the classification of histopathological melanoma images and thus shows promise to assist humanmelanoma diagnoses.
Authors: Juan C Vizcarra; Marla Gearing; Michael J Keiser; Jonathan D Glass; Brittany N Dugger; David A Gutman Journal: Acta Neuropathol Commun Date: 2020-04-28 Impact factor: 7.801
Authors: Max Schmitt; Roman Christoph Maron; Achim Hekler; Albrecht Stenzinger; Axel Hauschild; Michael Weichenthal; Markus Tiemann; Dieter Krahl; Heinz Kutzner; Jochen Sven Utikal; Sebastian Haferkamp; Jakob Nikolas Kather; Frederick Klauschen; Eva Krieghoff-Henning; Stefan Fröhling; Christof von Kalle; Titus Josef Brinker Journal: J Med Internet Res Date: 2021-02-02 Impact factor: 5.428
Authors: Jonas Scherer; Marco Nolden; Jens Kleesiek; Jasmin Metzger; Klaus Kades; Verena Schneider; Michael Bach; Oliver Sedlaczek; Andreas M Bucher; Thomas J Vogl; Frank Grünwald; Jens-Peter Kühn; Ralf-Thorsten Hoffmann; Jörg Kotzerke; Oliver Bethge; Lars Schimmöller; Gerald Antoch; Hans-Wilhelm Müller; Andreas Daul; Konstantin Nikolaou; Christian la Fougère; Wolfgang G Kunz; Michael Ingrisch; Balthasar Schachtner; Jens Ricke; Peter Bartenstein; Felix Nensa; Alexander Radbruch; Lale Umutlu; Michael Forsting; Robert Seifert; Ken Herrmann; Philipp Mayer; Hans-Ulrich Kauczor; Tobias Penzkofer; Bernd Hamm; Winfried Brenner; Roman Kloeckner; Christoph Düber; Mathias Schreckenberger; Rickmer Braren; Georgios Kaissis; Marcus Makowski; Matthias Eiber; Andrei Gafita; Rupert Trager; Wolfgang A Weber; Jakob Neubauer; Marco Reisert; Michael Bock; Fabian Bamberg; Jürgen Hennig; Philipp Tobias Meyer; Juri Ruf; Uwe Haberkorn; Stefan O Schoenberg; Tristan Kuder; Peter Neher; Ralf Floca; Heinz-Peter Schlemmer; Klaus Maier-Hein Journal: JCO Clin Cancer Inform Date: 2020-11
Authors: Sam Polesie; Phillip H McKee; Jerad M Gardner; Martin Gillstedt; Jan Siarov; Noora Neittaanmäki; John Paoli Journal: Front Med (Lausanne) Date: 2020-10-20