Comparative Analysis of Skin Cancer (Benign vs. Malignant) Detection Using Convolutional Neural Networks

We live in a world where people are suffering from many diseases. Cancer is the most threatening of them all. Among all the variants of cancer, skin cancer is spreading rapidly. It happens because of the abnormal growth of skin cells. The increase in ultraviolet radiation on the Earth’s surface is also helping skin cancer spread in every corner of the world. Benign and malignant types are the most common skin cancers people suffer from. People go through expensive and time-consuming treatments to cure skin cancer but yet fail to lower the mortality rate. To reduce the mortality rate, early detection of skin cancer in its incipient phase is helpful. In today’s world, deep learning is being used to detect diseases. The convolutional neural network (CNN) helps to find skin cancer through image classification more accurately. This research contains information about many CNN models and a comparison of their working processes for finding the best results. Pretrained models like VGG16, Support Vector Machine (SVM), ResNet50, and self-built models (sequential) are used to analyze the process of CNN models. These models work differently as there are variations in their layer numbers. Depending on their layers and work processes, some models work better than others. An image dataset of benign and malignant data has been taken from Kaggle. In this dataset, there are 6594 images of benign and malignant skin cancer. Using different approaches, we have gained accurate results for VGG16 (93.18%), SVM (83.48%), ResNet50 (84.39%), Sequential_Model_1 (74.24%), Sequential_Model_2 (77.00%), and Sequential_Model_3 (84.09%). This research compares these outcomes based on the model’s work process. Our comparison includes model layer numbers, working process, and precision. The VGG16 model has given us the highest accuracy of 93.18%.

Mutations of p53 gene in canine sweat gland carcinomas probably associated with UV radiation

Introduction Apocrine sweat gland carcinomas (ASGCs) are rare malignant skin tumours in dogs and humans. The literature published so far focuses mostly on the clinico-epidemiological aspect of these tumours, but little is known about their pathogenesis. In this study we aimed to determine whether the p53 gene is involved in the carcinogenesis of the apocrine sweat gland in dogs and whether ultraviolet radiation (UV) is related to it. Material and Methods Forty canine ASGCs were submitted to laser capture microdissection to isolate neoplastic cells, from which DNA was subsequently extracted. PCR amplification and sequencing of p53 exons 2–8 was then performed, followed by computer analysis of the obtained sequences. Results Sixteen mutations within the p53 gene were found in 13 tumours. The mutations involved C → T, T → C, G → A, and CC → TT transitions, C → G transversion and adenine deletion, which are gene alteration types known to be related to UV radiation in the process of skin carcinogenesis in humans. Six of the thirteen tumour cases displayed the C → T transitions in the same location in exon 4 and three of the thirteen cases displayed T → C in the same location in exon 5. Conclusion The results of the present study indicate both the participation of the p53 gene and the influence of UV radiation in the formation of ASGCs in dogs.

The state of cancer care in Russia: malignant neoplasms of the skin (C44). Typical mortality, median survival, observed and relative survival, taking into account the stage of the disease. Population-based research at the federal district level

In the previous article, we presented an analysis of the prevalence and quality of registration of patients with malignant neoplasms of the skin (C44) in Russia and the North-Western Federal District, showed the quality of accounting and revealed the features of the dynamics of localization and histological structure. In this report, for the first time in Russia, our attention is focused on the nature of evaluating the effectiveness of anti – cancer measures based on the main criterion-calculating the survival rates of patients with malignant skin neoplasms at the level of the North-Western Federal District, whose level of "rough" and standardized indicators of morbidity and mortality of the population practically does not differ from the average Russian one. Calculations of all types of survival rates of patients with malignant skin neoplasms (C44) at the federal district level became possible after the organization in February 2019 of a single database (DB) of the cancer registry of the North-Western Federal District, numbering more than 1 million cases of ZNO, including more than 20,000 with malignant melanoma of the skin (2.0%). The created database allows for in-depth development of the dynamics of the localization and histological structure for any part of the localization structure (lip, ear, trunk, etc. parts of the body), which is not available for conducting on the basis of state reporting forms formed before receiving all the information characterizing the state of the oncological service, not to mention that when analyzing the activities of the oncological service, databases of population cancer registers are not used and calculations of the main criterion for its assessment – the calculation of patient survival.

EP.WE.741Management of a giant extraocular sebaceous carcinoma during the COVID-19 Pandemic: a case report

Introduction Sebaceous carcinoma is a rare malignant skin tumour of the sebaceous glands which is most frequently located in the ocular region (75% of cases). Prompt recognition and treatment is vital, as it bears a high mortality rate ranging from 9 – 50%. We present a rare case of a giant sebaceous carcinoma of the upper back, managed as a day-surgery case because of resource and safety restrictions due to the COVID-19 pandemic. Case report We present a case of a 58-year-old male patient with a giant lesion on his upper back measuring 13 x 13cm in size. Due to the COVID-19 pandemic, this was resected under local anaesthetic. The histology and immunohistochemistry confirmed the diagnosis of a giant sebaceous carcinoma. Resection margins were clear and subsequent investigations confirmed there were no metastases. The defect was reconstructed with Matriderm © and a split-thickness skin graft, achieving a satisfactory cosmetic outcome at 3 months follow-up. Conclusion This is the first report which demonstrates the wide local excision under local anaesthetic for a rare presentation of a giant sebaceous carcinoma. This approach may be more widely utilised in the extraocular manifestation of this tumour, reducing the morbidity and length of hospital stay for such patients.

Bioinformatics Analysis Predicts ceRNA Regulatory Axes Related to Melanoma Pathogenesis

Background Melanoma is a highly malignant skin tumour, with an incidence and mortality rates accounting for approximately 5% and >75% of all tumours, respectively. In the present study, we aimed to screen mRNAs, microRNAs, and circRNAs related to the pathogenesis of melanoma via bioinformatics methods. Materials and methods The microarray data correlating with melanoma were obtained from the GEO database, and the differentially expressed genes between melanomatissues and non-melanoma tissues were screened using GEO2R. Moreover, the Hub genes were screened using the STRING database, Cytoscape software, and GEPIA database. The DAVID database was used for gene enrichment analysis. Thereafter, the ceRNA network diagram was constructed using the starBase database and was analysed for survival in order to predict the molecular markers for diagnosis and treatment of skin cutaneous melanoma; the most probable ceRNA mechanism was screened via pan-cancer analysis. Furthermore, the ceRNA network was verified using the GEPIA and UALCAN databases. Results In total, 266, 20, and 10 differentially expressed mRNAs, microRNAs, and circRNA10 were screened. Using Cytoscape software, 59 Hub and 31 key genes were screened, followed by construction of the ceRNA and ceRNA-gene enrichment analysis networks. Conclusion Clinical verification revealed that SIPA1L1 could regulate the expression of DSC3 by sponge adsorption of has-miR-106b and has-miR-20b, and thereby affect the pathogenesis and progression of skin cutaneous melanoma.