Classifying Skin Lesions: A Survey of Approaches in Skin Cancer Diagnosis

Skin cancer poses a substantial global health challenge, surpassing the combined incidence of colon, lung, and breast cancer cases in the United States. Timely detection is crucial, especially for melanoma, where early diagnosis significantly impacts patient outcomes. This study investigates the utilization of machine learning (ML) and deep learning (DL) methods to enhance skin cancer detection. We conduct a systematic comparison of various classification techniques and assess their efficacy. Our main goal is to critically assess current approaches and pinpoint areas for enhancement. This inquiry seeks to lay the groundwork for the development of advanced models that can streamline the diagnostic process, ultimately reducing the interval between skin cancer detection and treatment initiation.