• Michael A Marchetti, Konstantinos Liopyris, Stephen W Dusza, Noel C F Codella, David A Gutman, Brian Helba, Aadi Kalloo, Allan C Halpern, and International Skin Imaging Collaboration.
• Dermatology Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York. Electronic address: marchetm@mskcc.org.
• J. Am. Acad. Dermatol. 2020 Mar 1; 82 (3): 622-627.

BackgroundComputer vision has promise in image-based cutaneous melanoma diagnosis but clinical utility is uncertain.ObjectiveTo determine if computer algorithms from an international melanoma detection challenge can improve dermatologists' accuracy in diagnosing melanoma.MethodsIn this cross-sectional study, we used 150 dermoscopy images (50 melanomas, 50 nevi, 50 seborrheic keratoses) from the test dataset of a melanoma detection challenge, along with algorithm results from 23 teams. Eight dermatologists and 9 dermatology residents classified dermoscopic lesion images in an online reader study and provided their confidence level.ResultsThe top-ranked computer algorithm had an area under the receiver operating characteristic curve of 0.87, which was higher than that of the dermatologists (0.74) and residents (0.66) (P < .001 for all comparisons). At the dermatologists' overall sensitivity in classification of 76.0%, the algorithm had a superior specificity (85.0% vs. 72.6%, P = .001). Imputation of computer algorithm classifications into dermatologist evaluations with low confidence ratings (26.6% of evaluations) increased dermatologist sensitivity from 76.0% to 80.8% and specificity from 72.6% to 72.8%.LimitationsArtificial study setting lacking the full spectrum of skin lesions as well as clinical metadata.ConclusionAccumulating evidence suggests that deep neural networks can classify skin images of melanoma and its benign mimickers with high accuracy and potentially improve human performance.Copyright © 2019 American Academy of Dermatology, Inc. Published by Elsevier Inc. All rights reserved.

40 keywords...

hide…