Skin Cancer Detection Based on Deep Learning Network Architecture: An Analysis and Review

Skin cancer is very lethal and dangerous. It is due to the formation of mutations in skin cell DNA. If identified early and treated properly, the cancer can be managed; however, if it is not detected in time, it can go to different body parts. According to the estimates of 2022, there will be 7200 deaths due to malignant skin cancer. Computer aided diagnosis (CAD) is a no-pain procedure applied to identify the cancer automatically. This technique is composed of four distinct steps: accurate pre-processing, efficient lesion segmentation, significant feature extraction, and precise classification. In recent years, a lot of research has been conducted to apply machine learning, deep learning, and neural networks for classifying skin cancer. This review paper delivers an outline of the study done in this field, and grants the outcomes in the form of tables, techniques, and structures for improved understanding. Additionally, it emphasizes the advantages of using CAD for diagnosing skin cancer, such as cost-effectiveness and time-saving.