Literature DB >> 30852421

A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task.

Titus J Brinker1, Achim Hekler2, Alexander H Enk3, Joachim Klode4, Axel Hauschild5, Carola Berking6, Bastian Schilling7, Sebastian Haferkamp8, Dirk Schadendorf4, Stefan Fröhling2, Jochen S Utikal9, Christof von Kalle2.   

Abstract

BACKGROUND: Recent studies have demonstrated the use of convolutional neural networks (CNNs) to classify images of melanoma with accuracies comparable to those achieved by board-certified dermatologists. However, the performance of a CNN exclusively trained with dermoscopic images in a clinical image classification task in direct competition with a large number of dermatologists has not been measured to date. This study compares the performance of a convolutional neuronal network trained with dermoscopic images exclusively for identifying melanoma in clinical photographs with the manual grading of the same images by dermatologists.
METHODS: We compared automatic digital melanoma classification with the performance of 145 dermatologists of 12 German university hospitals. We used methods from enhanced deep learning to train a CNN with 12,378 open-source dermoscopic images. We used 100 clinical images to compare the performance of the CNN to that of the dermatologists. Dermatologists were compared with the deep neural network in terms of sensitivity, specificity and receiver operating characteristics.
FINDINGS: The mean sensitivity and specificity achieved by the dermatologists with clinical images was 89.4% (range: 55.0%-100%) and 64.4% (range: 22.5%-92.5%). At the same sensitivity, the CNN exhibited a mean specificity of 68.2% (range 47.5%-86.25%). Among the dermatologists, the attendings showed the highest mean sensitivity of 92.8% at a mean specificity of 57.7%. With the same high sensitivity of 92.8%, the CNN had a mean specificity of 61.1%.
INTERPRETATION: For the first time, dermatologist-level image classification was achieved on a clinical image classification task without training on clinical images. The CNN had a smaller variance of results indicating a higher robustness of computer vision compared with human assessment for dermatologic image classification tasks.
Copyright © 2019 The Authors. Published by Elsevier Ltd.. All rights reserved.

Entities:  

Keywords:  Artificial intelligence; Diagnostics; Melanoma; Skin cancer

Mesh:

Year:  2019        PMID: 30852421     DOI: 10.1016/j.ejca.2019.02.005

Source DB:  PubMed          Journal:  Eur J Cancer        ISSN: 0959-8049            Impact factor:   9.162


  42 in total

Review 1.  Designing deep learning studies in cancer diagnostics.

Authors:  Andreas Kleppe; Ole-Johan Skrede; Sepp De Raedt; Knut Liestøl; David J Kerr; Håvard E Danielsen
Journal:  Nat Rev Cancer       Date:  2021-01-29       Impact factor: 60.716

2.  Clinically Relevant Vulnerabilities of Deep Machine Learning Systems for Skin Cancer Diagnosis.

Authors:  Xinyi Du-Harpur; Callum Arthurs; Clarisse Ganier; Rick Woolf; Zainab Laftah; Manpreet Lakhan; Amr Salam; Bo Wan; Fiona M Watt; Nicholas M Luscombe; Magnus D Lynch
Journal:  J Invest Dermatol       Date:  2020-09-12       Impact factor: 8.551

Review 3.  The Enterprise Imaging Value Proposition.

Authors:  Cheryl A Petersilge
Journal:  J Digit Imaging       Date:  2020-02       Impact factor: 4.056

Review 4.  Artificial Intelligence: Review of Current and Future Applications in Medicine.

Authors:  L Brannon Thomas; Stephen M Mastorides; Narayan A Viswanadhan; Colleen E Jakey; Andrew A Borkowski
Journal:  Fed Pract       Date:  2021-11

5.  Keratinocytic Skin Cancer Detection on the Face Using Region-Based Convolutional Neural Network.

Authors:  Seung Seog Han; Ik Jun Moon; Woohyung Lim; In Suck Suh; Sam Yong Lee; Jung-Im Na; Seong Hwan Kim; Sung Eun Chang
Journal:  JAMA Dermatol       Date:  2020-01-01       Impact factor: 10.282

6.  Performance of a deep neural network in teledermatology: a single-centre prospective diagnostic study.

Authors:  C Muñoz-López; C Ramírez-Cornejo; M A Marchetti; S S Han; P Del Barrio-Díaz; A Jaque; P Uribe; D Majerson; M Curi; C Del Puerto; F Reyes-Baraona; R Meza-Romero; J Parra-Cares; P Araneda-Ortega; M Guzmán; R Millán-Apablaza; M Nuñez-Mora; K Liopyris; C Vera-Kellet; C Navarrete-Dechent
Journal:  J Eur Acad Dermatol Venereol       Date:  2020-11-22       Impact factor: 6.166

Review 7.  Machine learning for precision dermatology: Advances, opportunities, and outlook.

Authors:  Ernest Y Lee; Nolan J Maloney; Kyle Cheng; Daniel Q Bach
Journal:  J Am Acad Dermatol       Date:  2020-07-06       Impact factor: 11.527

8.  Toward automated assessment of mole similarity on dermoscopic images.

Authors:  Yao Zhang; Kamil Ali; Jacob A George; Jason S Reichenberg; Matthew C Fox; Adewole S Adamson; James W Tunnell; Mia K Markey
Journal:  J Med Imaging (Bellingham)       Date:  2021-02-10

Review 9.  The Importance of Incorporating Human Factors in the Design and Implementation of Artificial Intelligence for Skin Cancer Diagnosis in the Real World.

Authors:  Claire M Felmingham; Nikki R Adler; Zongyuan Ge; Rachael L Morton; Monika Janda; Victoria J Mar
Journal:  Am J Clin Dermatol       Date:  2021-03       Impact factor: 7.403

10.  Artificial intelligence (AI) in medicine as a strategic valuable tool.

Authors:  Andreas Larentzakis; Nik Lygeros
Journal:  Pan Afr Med J       Date:  2021-02-17
View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.