Pathologist-level classification of histopathological melanoma images with deep neural networks

TLDR

Pathologists diagnose most cancers by microscopic biopsy, yet inter‑observer discordance is high—about 25–26% for melanoma—though deep learning has improved accuracy in lung and breast cancer. This study aims to demonstrate deep learning’s potential to aid human assessment of histopathologic melanoma diagnosis. The authors scanned 695 H&E‑stained melanoma and nevus lesions, randomly cropped the images, and trained a convolutional neural network on 595 of them, testing its performance on 100 held‑out images. The CNN achieved discordance rates of 18% for melanoma, 20% for nevi, and 19% overall—comparable to human inter‑observer discordance—and demonstrated on‑par performance despite reduced data, time, and cost, indicating its value as an assistive tool.

Abstract

BackgroundThe 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 25–26% of discordance for classifying a benign nevus versus malignant melanoma. Deep learning was successfully implemented to enhance the precision of lung and breast cancer diagnoses. The aim of this study is to illustrate the potential of deep learning to assist human assessment for a histopathologic melanoma diagnosis.MethodsSix hundred ninety-five lesions were classified by an expert histopathologist in accordance with current guidelines (350 nevi and 345 melanomas). Only the haematoxylin and eosin stained (H&E) slides of these lesions were digitalised using a slide scanner and then randomly cropped. Five hundred ninety-five of the resulting images were used for the training of a convolutional neural network (CNN). The additional 100 H&E image sections were used to test the results of the CNN in comparison with the original class labels.FindingsThe total discordance with the histopathologist was 18% for melanoma (95% confidence interval [CI]: 7.4–28.6%), 20% for nevi (95% CI: 8.9–31.1%) and 19% for the full set of images (95% CI: 11.3–26.7%).InterpretationEven in the worst case, the discordance of the CNN was about the same compared with the discordance between human pathologists as reported in the literature. Despite the vastly reduced amount of data, time necessary for diagnosis and cost compared with the pathologist, our CNN archived on-par performance. Conclusively, CNNs indicate to be a valuable tool to assist human melanoma diagnoses.