The Epidemiology of Automatic Skin Cancer Detection by Comparative Analysis of Pre-processing and Segmentation Techniques

Abstract

Skin cancer is growing more widespread in many nations and is affecting a considerable number of individuals worldwide, including in Australia. Melanoma is a malignant skin change caused by melanocytes, which are pigment cells present in the epidermis (top layer of the skin). Early detection is strongly advised in order to effectively treat melanoma. As a result, the early excision of melanoma tissues has a significant impact on the survival rate of skin cancer patients. Traditional cancer detection methods are excruciatingly painful and time-consuming. As a result, to handle the many melanoma detection issues with high precision and accuracy, a quick and automated detection method is required. The procedure of identifying skin cancer by an automatic system is described in this work. Pre-processing is the most important of the numerous image processing phases. Its main goal is to provide a high-quality image by removing unnecessary portions and noise from the digital dermoscopy image. As a result, a detailed overview of several pre-processing methodologies is offered, as well as the work of scholars in this field who have primarily focused on pre-processing approaches.