Expert-Level Diagnosis of Nonpigmented Skin Cancer by Combined Convolutional Neural Networks

Abstract

IMPORTANCE Convolutional neural networks (CNNs) achieve expert-level accuracy in the diagnosis of pigmented melanocytic lesions. However, the most common types of skin cancer are nonpigmented and nonmelanocytic, and are more difficult to diagnose. OBJECTIVE To compare the accuracy of a CNN-based classifier with that of physicians with different levels of experience. DESIGN, SETTING, AND PARTICIPANTS A CNN-based classification model was trained on 7895 dermoscopic and 5829 close-up images of lesions excised at a primary skin cancer clinic between January 1, 2008, and July 13, 2017, for a combined evaluation of both imaging methods. The combined CNN (cCNN) was tested on a set of 2072 unknown cases and compared with results from 95 human raters who were medical personnel, including 62 board-certified dermatologists, with different experience in dermoscopy. MAIN OUTCOMES AND MEASURES The proportions of correct specific diagnoses and the accuracy to differentiate between benign and malignant lesions measured as an area under the receiver operating characteristic curve served as main outcome measures. RESULTS Among 95 human raters (51.6% female; mean age, 43.4 years; 95% CI, 41.0-45.7 years), the participants were divided into 3 groups (according to years of experience with dermoscopy): beginner raters (<3 years), intermediate raters (3-10 years), or expert raters (>10 years). The area under the receiver operating characteristic curve of the trained cCNN was higher than human ratings (0.742; 95% CI, 0.729-0.755 vs 0.695; 95% CI, 0.676-0.713; P < .001). The specificity was fixed at the mean level of human raters (51.3%), and therefore the sensitivity of the cCNN (80.5%; 95% CI, 79.0%-82.1%) was higher than that of human raters (77.6%; 95% CI, 74.7%-80.5%). The cCNN achieved a higher percentage of correct specific diagnoses compared with human raters (37.6%; 95% CI, 36.6%-38.4% vs 33.5%; 95% CI, 31.5%-35.6%; P = .001) but not compared with experts (37.3%; 95% CI, 35.7%-38.8% vs 40.0%; 95% CI, 37.0%-43.0%; P = .18). CONCLUSIONS AND RELEVANCE Neural networks are able to classify dermoscopic and close-up images of nonpigmented lesions as accurately as human experts in an experimental setting.

Machine Learning in Dermatology: Current Applications, Opportunities, and Limitations

“…In contrast, determining the diagnosis of melanoma versus benign pigmented nevus is a simpler binary classification problem. In 2019, Tschandl et al [52] demonstrated that a combined CNN (cCNN) can classify dermoscopic and clinical images of nonpigmented lesions on par with experts, namely 95 human raters of whom 62 were board-certified dermatologists. The authors combined the outputs of two CNNs, one trained with dermoscopic images and the other with clinical images.…”

Machine Learning for Clinical Decision-Making: Challenges and Opportunities

“…Another interesting application of DL is on predicting Alzheimer's disease approximately 6 years before the disease strikes, using PET brain scans (Ding et al, 2018). In dermatology, a few recent examples of DL showed higher accuracy than experts at diagnosing skin cancer (Esteva et al, 2017;Haenssle et al, 2018;Tschandl et al, 2019). In cardiology, a novel DL approach was implemented to automate diagnosis of acute ischemic infarction on CT (Beecy et al, 2018).…”

Use of artificial intelligence in skin cancer diagnosis and management

“…Several studies have now shown that CNNs trained on retrospective image data collected at a single time point are capable of classifying skin cancer with sensitivities and specificities equal or superior to that of dermatologists (Box 1), and clinicians with less experience gain most from AI support under experimental conditions . Hypomelanotic and acral melanoma can be more challenging to diagnose clinically, and this could potentially present a challenge for automated classification.…”

“…Hypomelanotic and acral melanoma can be more challenging to diagnose clinically, and this could potentially present a challenge for automated classification. However, CNNs have achieved greater accuracy for hypopigmented and acral lesions in comparison with human experts, at least in silica . In addition to clinical images, CNNs have been applied to histopathological images of melanoma and benign naevi with promising results …”