Expression Profiles of ASIC1/2 and TRPV1/4 in Common Skin Tumors.

The acid-sensing ion channels ASIC1 and ASIC2, as well as the transient receptor potential vanilloid channels TRPV1 and TRPV4, are proton-gated cation channels that can be activated by low extracellular pH (pHe), which is a hallmark of the tumor microenvironment in solid tumors. However, the role of these channels in the development of skin tumors is still unclear. In this study, we investigated the expression profiles of ASIC1, ASIC2, TRPV1 and TRPV4 in malignant melanoma (MM), squamous cell carcinoma (SCC), basal cell carcinoma (BCC) and in nevus cell nevi (NCN). We conducted immunohistochemistry using paraffin-embedded tissue samples from patients and found that most skin tumors express ASIC1/2 and TRPV1/4. Striking results were that BCCs are often negative for ASIC2, while nearly all SCCs express this marker. Epidermal MM sometimes seem to lack ASIC1 in contrast to NCN. Dermal portions of MM show strong expression of TRPV1 more frequently than dermal NCN portions. Some NCN show a decreasing ASIC1/2 expression in deeper dermal tumor tissue, while MM seem to not lose ASIC1/2 in deeper dermal portions. ASIC1, ASIC2, TRPV1 and TRPV4 in skin tumors might be involved in tumor progression, thus being potential diagnostic and therapeutic targets.

Combining CNN-based histologic whole slide image analysis and patient data to improve skin cancer classification.

BACKGROUND: Clinicians and pathologists traditionally use patient data in addition to clinical examination to support their diagnoses. OBJECTIVES: We investigated whether a combination of histologic whole slides image (WSI) analysis based on convolutional neural networks (CNNs) and commonly available patient data (age, sex and anatomical site of the lesion) in a binary melanoma/nevus classification task could increase the performance compared with CNNs alone. METHODS: We used 431 WSIs from two different laboratories and analysed the performance of classifiers that used the image or patient data individually or three common fusion techniques. Furthermore, we tested a naive combination of patient data and an image classifier: for cases interpreted as 'uncertain' (CNN output score <0.7), the decision of the CNN was replaced by the decision of the patient data classifier. RESULTS: The CNN on its own achieved the best performance (mean ± standard deviation of five individual runs) with AUROC of 92.30% ± 0.23% and balanced accuracy of 83.17% ± 0.38%. While the classification performance was not significantly improved in general by any of the tested fusions, naive strategy of replacing the image classifier with the patient data classifier on slides with low output scores improved balanced accuracy to 86.72% ± 0.36%. CONCLUSION: In most cases, the CNN on its own was so accurate that patient data integration did not provide any benefit. However, incorporating patient data for lesions that were classified by the CNN with low 'confidence' improved balanced accuracy.

Deep neural networks are superior to dermatologists in melanoma image classification.

BACKGROUND: Melanoma is the most dangerous type of skin cancer but is curable if detected early. Recent publications demonstrated that artificial intelligence is capable in classifying images of benign nevi and melanoma with dermatologist-level precision. However, a statistically significant improvement compared with dermatologist classification has not been reported to date. METHODS: For this comparative study, 4204 biopsy-proven images of melanoma and nevi (1:1) were used for the training of a convolutional neural network (CNN). New techniques of deep learning were integrated. For the experiment, an additional 804 biopsy-proven dermoscopic images of melanoma and nevi (1:1) were randomly presented to dermatologists of nine German university hospitals, who evaluated the quality of each image and stated their recommended treatment (19,296 recommendations in total). Three McNemar's tests comparing the results of the CNN's test runs in terms of sensitivity, specificity and overall correctness were predefined as the main outcomes. FINDINGS: The respective sensitivity and specificity of lesion classification by the dermatologists were 67.2% (95% confidence interval [CI]: 62.6%-71.7%) and 62.2% (95% CI: 57.6%-66.9%). In comparison, the trained CNN achieved a higher sensitivity of 82.3% (95% CI: 78.3%-85.7%) and a higher specificity of 77.9% (95% CI: 73.8%-81.8%). The three McNemar's tests in 2 × 2 tables all reached a significance level of p < 0.001. This significance level was sustained for both subgroups. INTERPRETATION: For the first time, automated dermoscopic melanoma image classification was shown to be significantly superior to both junior and board-certified dermatologists (p < 0.001).

Deep learning outperformed 11 pathologists in the classification of histopathological melanoma images.

BACKGROUND: The 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 on 25-26% of discordance for classifying a benign nevus versus malignant melanoma. A recent study indicated the potential of deep learning to lower these discordances. However, the performance of deep learning in classifying histopathologic melanoma images was never compared directly to human experts. The aim of this study is to perform such a first direct comparison. METHODS: A total of 695 lesions were classified by an expert histopathologist in accordance with current guidelines (350 nevi/345 melanoma). Only the haematoxylin & eosin (H&E) slides of these lesions were digitalised via a slide scanner and then randomly cropped. A total of 595 of the resulting images were used to train a convolutional neural network (CNN). The additional 100 H&E image sections were used to test the results of the CNN in comparison to 11 histopathologists. Three combined McNemar tests comparing the results of the CNNs test runs in terms of sensitivity, specificity and accuracy were predefined to test for significance (p < 0.05). FINDINGS: The CNN achieved a mean sensitivity/specificity/accuracy of 76%/60%/68% over 11 test runs. In comparison, the 11 pathologists achieved a mean sensitivity/specificity/accuracy of 51.8%/66.5%/59.2%. Thus, the CNN was significantly (p = 0.016) superior in classifying the cropped images. INTERPRETATION: With limited image information available, a CNN was able to outperform 11 histopathologists in the classification of histopathological melanoma images and thus shows promise to assist human melanoma diagnoses.

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

BACKGROUND: The 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. METHODS: Six 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. FINDINGS: The 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%). INTERPRETATION: Even 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.

Classification of skin lesions using transfer learning and augmentation with Alex-net.

Skin cancer is one of most deadly diseases in humans. According to the high similarity between melanoma and nevus lesions, physicians take much more time to investigate these lesions. The automated classification of skin lesions will save effort, time and human life. The purpose of this paper is to present an automatic skin lesions classification system with higher classification rate using the theory of transfer learning and the pre-trained deep neural network. The transfer learning has been applied to the Alex-net in different ways, including fine-tuning the weights of the architecture, replacing the classification layer with a softmax layer that works with two or three kinds of skin lesions, and augmenting dataset by fixed and random rotation angles. The new softmax layer has the ability to classify the segmented color image lesions into melanoma and nevus or into melanoma, seborrheic keratosis, and nevus. The three well-known datasets, MED-NODE, Derm (IS & Quest) and ISIC, are used in testing and verifying the proposed method. The proposed DCNN weights have been fine-tuned using the training and testing dataset from ISIC in addition to 10-fold cross validation for MED-NODE and DermIS-DermQuest. The accuracy, sensitivity, specificity, and precision measures are used to evaluate the performance of the proposed method and the existing methods. For the datasets, MED-NODE, Derm (IS & Quest) and ISIC, the proposed method has achieved accuracy percentages of 96.86%, 97.70%, and 95.91% respectively. The performance of the proposed method has outperformed the performance of the existing classification methods of skin cancer.

Towards Interpretable Skin Lesion Classification with Deep Learning Models.

Skin disease is a prevalent condition all over the world. Computer vision-based technology for automatic skin lesion classification holds great promise as an effective screening tool for early diagnosis. In this paper, we propose an accurate and interpretable deep learning pipeline to achieve such a goal. Comparing with existing research, we would like to highlight the following aspects of our model. 1) Rather than a single model, our approach ensembles a set of deep learning architectures to achieve better classification accuracy; 2) Generative adversarial network (GAN) is involved in the model training to promote data scale and diversity; 3) Local interpretable model-agnostic explanation (LIME) strategy is applied to extract evidence from the skin images to support the classification results. Our experimental results on real-world skin image corpus demonstrate the effectiveness and robustness of our method. The explainability of our model further enhances its applicability in real clinical practice.

Epidermal Nevi and Related Syndromes -- Part 1: Keratinocytic Nevi. / Nevus epidérmicos y síndromes relacionados. Parte 1: nevus queratinocíticos.

Epidermal nevi are hamartomatous lesions derived from the epidermis and/or adnexal structures of the skin; they have traditionally been classified according to their morphology. New variants have been described in recent years and advances in genetics have contributed to better characterization of these lesions and an improved understanding of their relationship with certain extracutaneous manifestations. In the first part of this review article, we will look at nevi derived specifically from the epidermis and associated syndromes.

Epidermal Nevi and Related Syndromes - Part 2: Nevi Derived from Adnexal Structures. / Nevus epidérmicos y síndromes relacionados. Parte 2: Nevus derivados de estructuras anexiales.

Epidermal nevi are hamartomatous lesions derived from the epidermis and/or adnexal structures of the skin; they have traditionally been classified according to their morphology. New variants have been described in recent years and advances in genetics have contributed to better characterization of these lesions and an improved understanding of their relationship with certain extracutaneous manifestations. In the second part of this review article, we will look at nevi derived from the adnexal structures of the skin and associated syndromes.

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.

Biologically distinct subsets of nevi.

Melanocytic nevi (MN) encompass a range of benign tumors with varying microscopic and macroscopic features. Their development is a multifactorial process under genetic and environmental influences. The clinical importance of MN lies in distinguishing them from melanoma and in recognizing their associations with melanoma risk and cancer syndromes. Historically, the distinction between the different types of MN, as well as between MN and melanoma, was based on clinical history, gross morphology, and histopathological features. While histopathology with clinical correlation remains the gold standard for differentiating and diagnosing melanocytic lesions, in some cases, this may not be possible. The use of dermoscopy has allowed for the assessment of subsurface skin structures and has contributed to the clinical evaluation and classification of MN. Genetic profiling, while still in its early stages, has the greatest potential to refine the classification of MN by clarifying their developmental processes, biological behaviors, and relationships to melanoma. Here we review the most salient clinical, dermoscopic, histopathological, and genetic features of different MN subgroups.

Klippel-Trenaunay syndrome belongs to the PIK3CA-related overgrowth spectrum (PROS).

Klippel-Trenaunay syndrome (KTS), originally described as a triad of cutaneous capillary malformation, bone and soft-tissue hypertrophy, as well as venous and lymphatic malformations, has been considered by dermatologists as a distinct diagnostic entity. However, cases with KTS have also been reported to have neurological disorders, developmental delay and digital abnormalities, indicating multisystem involvement. Recently, a number of overgrowth syndromes, with overlapping phenotypic features with KTS, have been identified; these include MCAP and CLOVES syndromes as well as fibroadipose hyperplasia. These conditions harbour mutations in the PIK3CA gene, and they have been included in the PIK3CA-related overgrowth spectrum (PROS). Based on recent demonstrations of PIK3CA mutations also in KTS, it appears that, rather than being a distinct diagnostic entity, KTS belongs to PROS. These observations have potential diagnostic and therapeutic implications for KTS.

Hybrid eccrine gland and hair follicle hamartoma: a new entity of adnexal nevus.

Eccrine nevus shows increase in number or size of eccrine glands, whereas hair follicle nevus is composed of densely packed normal vellus hairs, and eccrine-pilar angiomatous nevus reveals increase of eccrine, pilar, and angiomatous structures. No case with increased number of both eccrine glands and hair follicles only in the dermis has been previously reported. A 10-month-old girl presented with cutaneous hamartoma with overlying skin hyperpigmentation on her left hypochondrium since 3 months of age, in whom the lesion was completely excised. Histopathology demonstrated evidently increased number of both eccrine glands and hair follicles in the dermis with reactive hyperplasia of collagen fibers. No recurrence occurred after the tumor was completely excised. A term "hybrid eccrine gland and hair follicle hamartoma" is proposed for this unique lesion.

Nevus and melanoma paraconsistent classification.

This paper presents the first studies on Nevus and Melanoma classification by using Paraconsistent Artificial Neural Network (PANN). Nevus is usually a small growth on the skin while Melanoma is a dangerous skin cancer. The proposed automated process classifies a set of medical images as Nevus and Melanoma based on a methodology grounded on PANN which is able to deal with conflicting, paracomplete and imprecise data directly without trivialization. Such methodology performed promising results considering only border features to classify the sample.

[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.

Nevus anelasticus: how should such lesions be classified?

Nevus anelasticus represents a rare entity that is most commonly classified as a connective tissue nevus. It typically presents before 20 years of age with asymmetrically distributed white-to-skin-toned or pink-to-red papules or plaques on the trunk and upper extremities. The lesion is defined histopathologically by the absence or degeneration of elastic fibers in the dermis. We report the case of a healthy 17-year-old female who presented with an asymptomatic slowly progressive plaque on the right inferior areola. Histopathologic examination showed the absence of elastic fibers in the papillary and upper reticular dermis and fragmented elastic tissue fibers in the deep reticular dermis. Although there is ongoing controversy regarding the nosology of this uncommon disorder, we propose that it is a distinct entity based on its histopathologic and clinical features.

Proposal for a clinical-dermoscopic classification of scalp naevi.

BACKGROUND: Recent research suggests that scalp naevi differ with respect to their epidemiology, patient characteristics and morphological patterns, but currently a classification of scalp naevi is lacking. OBJECTIVES: To investigate the prevalence, together with clinical and dermoscopic features, of scalp naevi detected in persons attending a skin cancer screening programme, and to elaborate a classification of scalp naevi based on their most common morphological patterns. METHODS: Participants were recruited during the melanoma prevention programme 'sun watch' of Austrian Cancer Aid in Styria. Each participant received a clinical and dermoscopic total-body skin examination including the scalp. For each participant, demographics and clinical characteristics including number of scalp naevi were recorded. Clinical and dermoscopic photographs of at least one scalp naevus per participant were taken and evaluated for specific clinical and dermoscopic features. RESULTS: In total 867 subjects, including 119 participants (13·7%) with scalp naevi, participated in the study. Compared with those without scalp naevi, subjects with scalp naevi were significantly younger, were more often men and more often exhibited congenital naevi on the body (P < 0·01 for all). Analysis of the clinical and dermoscopic variability of scalp naevi allowed for a proposal to classify scalp naevi into six main groups, namely common, papillomatous, eclipse, congenital, blue and atypical naevus. CONCLUSIONS: Scalp naevi can be classified into six morphological groups; scalp lesions deviating from these six main patterns should be carefully managed to rule out melanoma.