Abstract
Skin cancer, a widespread and potentially life-threatening ailment affecting people globally, underscores the significance of early detection. Early identification significantly reduces treatment duration, costs, and patient suffering compared to traditional treatments like chemotherapy. This review focuses on skin cancer detection's crucial role in healthcare, offering an extensive examination of research papers from 2018 to 2022. Deep learning methods, explored in these papers, emerge as robust tools automating and improving skin cancer detection accuracy, thereby lightening the load on medical professionals. The review spans diverse deep learning approaches—convolutional neural networks, transfer learning, and ensemble models. We scrutinize the use of various datasets and data augmentation strategies in model training. The presented findings aim to contribute to more timely, accurate, and accessible diagnoses in the ongoing battle against skin cancer.