In this study, the authors discuss whether the use of deep learning convolutional neural networks, or DL-CNNs, in coordination with standard dermatological practices increases accuracy when diagnosing cutaneous melanoma. The study was performed by comparing the diagnostic performance of 27 dermatologists to that of the DL-CNN when given 200 lesions. Some dermatologists were given access to the DL-CNN data to see if the information improved their diagnostic accuracy compared to those without it. Each group was then measured for specificity, sensitivity, and ROC-AUC. Vollmer found that the DL-CNN was comparable in sensitivity, but inferior in specificity and ROC-AUC to the dermatologists without access to the DL-CNN data. When given the DL-CNN data, the dermatologists’ sensitivity increased further, without decreasing their ROC-AUC or specificity. The DL-CNN was able to diagnose many types of cutaneous melanoma at a higher rate than the dermatologists, particularly lentigo maligna, nodular cutaneous melanoma, and OTH-M. When dermatologists were given DL-CNN data to assist diagnosis, their diagnostic sensitivity increased variably, with a notable 22.3% increase in identifying OTH-M.