Diagnostic performance of artificial intelligence for histologic melanoma recognition compared to 18 international expert pathologists
Autor/a
Brinker, Titus J.
Schmitt, Max
Krieghoff-Henning, Eva I.
Barnhill, Raymond L.
Beltraminelli, Helmut
Braun, Stephan A.
Carr, Richard
Fernandez Figueras, Maria-Teresa
Ferrara, Gerardo
Fraitag, Sylvie
Gianotti, Raffaele
Llamas-Velasco, Mar
Müller, Cornelia S.L.
Perasole, Antonio
Requena, Luis
Sangueza, Omar P.
Santonja, Carlos
Starz, Hans
Vale, Esmeralda
Weyers, Wolfgang
Dipl-Inform, Achim Hekler
Kather, Jakob N.
Fröhling, Stefan
Krahl, Dieter
Holland-Letz, Tim
Utikal, Jochen S.
Saggini, Andrea
Kutzner, Heinz
Fecha de publicación
2021-02ISSN
0190-9622
Resumen
To the Editor: Currently, pathologic melanoma classification is based on the—inevitably somewhat subjective—integration of several histologic features.1 Thus, discordance between pathologists classifying the same lesions can be substantial, and objective assistance tools are needed. The classification of dermoscopic skin lesion images based on convolutional neural networks (CNNs) works well.2 On a histologic level, our pilot studies provided a proof regarding the principle of CNN-based melanoma recognition using tiny sections of hematoxylin-eosin–stained digitized slides. We compared the ability of CNNs with that of 18 international expert pathologists from eight different countries to discriminate melanomas and nevi in a less artificial setting using hematoxylin-eosin–stained whole-slide images. Ensembles of 3 individual CNNs were trained and tested using single hematoxylin-eosin–stained whole-slide images of 50 individual melanomas and 50 nevi labeled by a panel of 2 experienced dermatopathologists according to the standard practice to provide the “ground truth” (Supplementary Figs 1 and 2 available via Mendeley at https://data.mendeley.com/datasets/j87c9jshxy/1, Supplementary Table I available via Mendeley at https://data.mendeley.com/datasets/j87c9jshxy/1). The same 100 digitized slides were diagnosed using a web-based survey by 18 international dermatopathologists, each with at least 5 years of experience.
Tipo de documento
Artículo
Versión del documento
Versión publicada
Lengua
Inglés
Materias (CDU)
61 - Medicina
Palabras clave
Intel·ligència artificial
Melanoma
Microorganismes patògens
Inteligencia artificial
Melanoma
Microorganismos
Artificial intelligence
Melanoma
Pathogenic microorganisms
Páginas
3
Publicado por
Elsevier
Publicado en
Journal of the American Academy of Dermatology
Citación
Brinker, Titus J.; Schmitt, Max; Krieghoff-Henning, Eva I. [et al.]. Diagnostic performance of artificial intelligence for histologic melanoma recognition compared to 18 international expert pathologists. Journal of the American Academy of Dermatology, 2021, p. 1-3. Disponible en: <https://www.sciencedirect.com/science/article/pii/S0190962221003315?via%3Dihub>. Fecha de acceso: 23 mar. 2021. DOI: 10.1016/j.jaad.2021.02.009
Este ítem aparece en la(s) siguiente(s) colección(ones)
- Ciències de la Salut [725]
Derechos
© 2021 by the American Academy of Dermatology, Inc.This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Excepto si se señala otra cosa, la licencia del ítem se describe como http://creativecommons.org/licenses/by/4.0/