Prior detection of Melanoma skin cancer using Deep Learning

Anomalous growth of skin cells caused due to harmful Sun’s rays is predominant form of cancer. According to WHO skin cancer is the fifth most common cancer across the world. In 2020, 324635 new cases were accounted of both sexes world wide & 57043 deaths have been reported by Globocan 2020. Skin cancer is categorized in to three: Basal Cell carcinoma, Melanoma & Squamous Cell Carcinoma. Furthermore Melanoma is the most lethal malignancies in the world with very less chances of Survival. Early detection of skin cancer is crucial as it provides treatment options & better survival rates, conversely it may spread to other organs & can access lymphatic vessels. Assuming the danger of skin cancer, researchers have developed many early detection methods. Colour, symmetry, form, size, & other lesion criteria are utilised to detect & distinguish between benign & malignant skin cancers. Convolutional neural networks (CNN) are one of the most effective approaches for diagnosing melanoma today. This paper provides an overview of deep learning algorithms for the early diagnosis of skin cancer. The topic of skin cancer diagnosis-related research articles from credible publishers was investigated. The table shows a comparative examination of skin cancer detection using deep learning algorithms.