Dermatologist-level classification of skin cancer with deep neural networks.

Skin cancer, the most common human malignancy, is primarily diagnosed visually, beginning with an initial clinical screening and followed potentially by dermoscopic analysis, a biopsy and histopathological examination. Automated classification of skin lesions using images is a challenging task owing to the fine-grained variability in the appearance of skin lesions. Deep convolutional neural networks (CNNs) show potential for general and highly variable tasks across many fine-grained object categories. Here we demonstrate classification of skin lesions using a single CNN, trained end-to-end from images directly, using only pixels and disease labels as inputs. We train a CNN using a dataset of 129,450 clinical images-two orders of magnitude larger than previous datasets-consisting of 2,032 different diseases. We test its performance against 21 board-certified dermatologists on biopsy-proven clinical images with two critical binary classification use cases: keratinocyte carcinomas versus benign seborrheic keratoses; and malignant melanomas versus benign nevi. The first case represents the identification of the most common cancers, the second represents the identification of the deadliest skin cancer. The CNN achieves performance on par with all tested experts across both tasks, demonstrating an artificial intelligence capable of classifying skin cancer with a level of competence comparable to dermatologists. Outfitted with deep neural networks, mobile devices can potentially extend the reach of dermatologists outside of the clinic. It is projected that 6.3 billion smartphone subscriptions will exist by the year 2021 (ref. 13) and can therefore potentially provide low-cost universal access to vital diagnostic care.

[Genetic basis of seborrheic keratosis and epidermal nevi]. / Genetische Grundlagen seborrhoischer Keratosen und epidermaler Nävi.

Seborrheic keratosis (SK) and epidermal nevi (EN) represent benign skin tumors and congenital lesions, respectively. Oncogenic mutations are fundamentally involved in their pathogenesis and SK is characterized by a broad spectrum of somatic mutations in the FGFR3, PIK3CA, RAS, AKT1 and EGFR genes. In contrast to malignant tumors, SK is genetically stable without alterations of tumor suppressor genes. The ENs are caused by postzygotic activating hot spot mutations in FGFR3, PIK3CA and particularly HRAS, resulting in a genetic mosaicism. The size of the lesions and the differentiation potential of the mutated cell into various tissue types depends on the time point of the mutation during embryogenesis. The genetic mosaic may predispose to a later growth of benign and malignant (adnexal) tumors.

Epithelial keratin and filaggrin expression in seborrheic keratosis: evaluation based on histopathological classification.

OBJECTIVES: Seborrheic keratosis (SK) is classified into six types: hyperkeratotic; acanthotic; irritated; clonal; reticulated; and adenoid. However, the origins of the respective types of SK remain unclear. METHODS: To clarify the histogenetic origins of SK, we performed immunohistochemical studies of keratin (K) and filaggrin expression, taking into account the histopathological classifications of SK. RESULTS: Hyperkeratotic SK mainly expressed K1, K10, and filaggrin. Acanthotic SK mainly expressed K14 with some K15. Irritated SK mainly expressed K14 and K17 in squamous eddies. Clonal SK, reticulated SK, and adenoid SK mainly expressed K14. The results show that hyperkeratotic SK differentiated towards squamoid terminal keratinization, whereas acanthotic, irritated, clonal, reticulated, and adenoid SK mainly differentiated towards basaloid undifferentiated cells. In addition, acanthotic SK differentiated towards the hair bulge, and irritated SK differentiated towards the follicular infrainfundibulum. CONCLUSIONS: Based on the patterns of keratin and filaggrin expression demonstrated by the histopathological types, SK demonstrated diverse differentiation towards epidermal keratinization, basaloid cells, the infrainfundibulum and hair follicle bulges, which suggests that SK is in an undifferentiated and hyperproliferative state with heterogeneous differentiation. The immunohistochemical method of investigating patterns of keratin expression is useful in the differential diagnosis of cutaneous epithelial tumors.

Are all seborrheic keratoses benign? Review of the typical lesion and its variants.

BACKGROUND: Seborrheic keratosis (SK) is one of the more common benign epidermal neoplasms seen in adult and middle-aged patients. OBJECTIVE: As little is written in the literature about the variants of SK, this article aims to categorize and discuss the different subtypes and their important associations. METHODS: An in-depth literature search using OVID Medline and PubMed was conducted to classify the various subtypes of SK. Clinical variants were photographed and used to help document the subtypes. The pathology is described for each. RESULTS: Six subtypes of SK were identified: dermatosis papulosa nigra, stucco keratosis, inverted follicular keratosis, large cell acanthoma, lichenoid keratosis, and flat seborrheic keratosis. Although the etiology and pathogenesis of SKs are still largely debatable, several underlying mechanisms and contributing factors have been identified. All subtypes represent benign lesions, and treatment is usually done for cosmetic reasons. Several of the subtypes may act as cutaneous markers for internal malignancy and should be monitored closely for any atypical changes. CONCLUSION: Although all subtypes of SK are benign, their association with other malignant lesions and ability to serve as cutaneous markers of internal malignancy emphasize the importance of correctly identifying all variants.

Seborrheic keratosis with pseudorosettes and adamantinoid seborrheic keratosis: two new histopathologic variants.

AIMS: Seborrheic keratoses are the most common benign cutaneous neoplasms in adult and middle-age patients. There are six distinctive histopathologic variants of seborrheic keratosis, namely, acanthotic or solid, reticulated or adenoid, hyperkeratotic or papillomatous, clonal or nested, irritated and inflamed. METHODS: We report two additional histopathologic variants of seborrheic keratosis. RESULTS: One lesion showed abundant intercellular mucin, closely resembling to adamantinoma, and therefore was named adamantinoid seborrheic keratosis. The other one exhibited a peculiar distribution of the basaloid keratinocytes, which were arranged radially around small central spaces, resulting in pseudorosette formation. DISCUSSION: To our knowledge, these two histopathologic variants of seborrheic keratosis have been not previously described in the literature.

Queratosis seborreica vs melanoma: ¿La dermatoscopía es útil en el diagnóstico diferencial? / Seborrhoeic keratoses versus melanoma: can dermoscopy help?

Las queratosis seborreicas (QS) son tumores benignos de la piel que aparecen luego de la edad media de la vida y en ocasiones pueden verse en personas jóvenes. Generalmente el diagnóstico clínico de las QS no ofrece dificultad. Sin embargo las formas superficiales pigmentadas se deben diferenciar del lentigo maligno melanoma, mientras que las formas sobreelevadas se pueden confundir con nevos atípicos y melanoma. Es en éstos casos dudosos en los que la dermatoscopía mejora el diagnóstico clínico más que en ninguna otra lesión pigmentada exceptuando al hemangioma (AU)