• Michael A Marchetti, Noel C F Codella, Stephen W Dusza, David A Gutman, Brian Helba, Aadi Kalloo, Nabin Mishra, Cristina Carrera, M Emre Celebi, Jennifer L DeFazio, Natalia Jaimes, Ashfaq A Marghoob, Elizabeth Quigley, Alon Scope, Oriol Yélamos, Allan C Halpern, and International Skin Imaging Collaboration.
• Dermatology Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York.
• J. Am. Acad. Dermatol. 2018 Feb 1; 78 (2): 270-277.e1.

BackgroundComputer vision may aid in melanoma detection.ObjectiveWe sought to compare melanoma diagnostic accuracy of computer algorithms to dermatologists using dermoscopic images.MethodsWe conducted a cross-sectional study using 100 randomly selected dermoscopic images (50 melanomas, 44 nevi, and 6 lentigines) from an international computer vision melanoma challenge dataset (n = 379), along with individual algorithm results from 25 teams. We used 5 methods (nonlearned and machine learning) to combine individual automated predictions into "fusion" algorithms. In a companion study, 8 dermatologists classified the lesions in the 100 images as either benign or malignant.ResultsThe average sensitivity and specificity of dermatologists in classification was 82% and 59%. At 82% sensitivity, dermatologist specificity was similar to the top challenge algorithm (59% vs. 62%, P = .68) but lower than the best-performing fusion algorithm (59% vs. 76%, P = .02). Receiver operating characteristic area of the top fusion algorithm was greater than the mean receiver operating characteristic area of dermatologists (0.86 vs. 0.71, P = .001).LimitationsThe dataset lacked the full spectrum of skin lesions encountered in clinical practice, particularly banal lesions. Readers and algorithms were not provided clinical data (eg, age or lesion history/symptoms). Results obtained using our study design cannot be extrapolated to clinical practice.ConclusionDeep learning computer vision systems classified melanoma dermoscopy images with accuracy that exceeded some but not all dermatologists.Copyright © 2017 American Academy of Dermatology, Inc. Published by Elsevier Inc. All rights reserved.

23 keywords...

hide…