Although there have been reports of the successful diagnosis of skin disorders using deep learning, unrealistically large clinical image datasets are required for artificial intelligence (AI) training. We created datasets of standardized nail images using a region-based convolutional neural network (R-CNN) trained to distinguish the nail from the background. We used R-CNN to generate training datasets of 49,567 images, which we then used to fine-tune the ResNet-152 and VGG-19 models. The validation datasets comprised 100 and 194 images from Inje University (B1 and B2 datasets, respectively), 125 images from Hallym University (C dataset), and 939 images from Seoul National University (D dataset). The AI (ensemble model; ResNet-152 + VGG-19 + feedforward neural networks) results showed test sensitivity/specificity/ area under the curve values of (96.0 / 94.7 / 0.98), (82.7 / 96.7 / 0.95), (92.3 / 79.3 / 0.93), (87.7 / 69.3 / 0.82) for the B1, B2, C, and D datasets. With a combination of the B1 and C datasets, the AI Youden index was significantly (p = 0.01) higher than that of 42 dermatologists doing the same assessment manually. For B1+C and B2+ D dataset combinations, almost none of the dermatologists performed as well as the AI. By training with a dataset comprising 49,567 images, we achieved a diagnostic accuracy for onychomycosis using deep learning that was superior to that of most of the dermatologists who participated in this study.

Convolutional neural network (CNN) has shown expert-level performance in the fields of ophthalmology, dermatology, and radiology.^1-9 Convolutional neural networks showed successful results in ophthalmologic emergency determination,¹⁰ brain hemorrhage detection with computed tomographic scan,⁴ and multiple-class cardiopulmonary disease classification with radiographic images.⁷ In ophthalmology, the IDx-DR diagnostic system (IDx Technologies Inc)¹¹ was approved by the US Food and Drug Administration to be used independently in screening for diabetic retinopathy.

Regulation of melanogenesis has been the focus of treatment for hyperpigmentary skin disorders. Although hesperidin is one of the most well-known, naturally occurring flavonoids with antioxidant and anti-inflammatory effect, its anti-melanogenic effect is not known. The present study aims to determine the anti-melanogenic effect of hespiridin as well as its underlying molecular mechanisms. Melanin contents were measured in normal human melanocytes and B16F10 melanoma cells. Protein and mRNA levels of tyrosinase, microphthalmia-associated transcription factor (MITF), tyrosinase related protein-1 (TRP-1) and TRP-2 were determined. Melanogenesis-regulating signals were examined. In results, hesperidin strongly inhibited melanin synthesis and tyrosinase activity. Hesperidin decreased tyrosinase, TRP-1, and TRP-2 protein expression but increased phospho-extracellular signal-regulated kinase 1/2 (p-Erk1/2) expression. Specific inhibitor of Erk1/2 or proteasome inhibitor reversed the inhibition of melanogenesis induced by hesperidin. Taken together, hesperidin, a popular antioxidant, stimulated Erk1/2 phosphorylation which subsequently degraded MITF which resulted in suppression of melanogenic enzymes and melanin synthesis.