Initial Prediction of Skin Cancer Using Deep Learning Techniques: A Systematic Review

Abstract

Skin disease is a common medical condition that affects the outer layer of our body and requires early intervention to prevent it from becoming life-threatening. The techniques of deep learning have been developed as a necessary tool for identifying skin diseases, attracting the attention of researchers. In this review, we examine the efforts of researchers who have utilized deep learning technology for skin disease identification. We provide an overview of skin diseases, including their types, datasets, and data pre-processing techniques. Furthermore, we explore deep learning approaches and popular methods used by researchers in diagnosing skin diseases. The primary aim of this study is to present acomprehensive review of recent research on skin disease detection using deep learning methodologies. Our observations demonstrate that these methods, known for their accuracy, outperform dermatologists, machine-based therapies, and other classification methods in recognizing skin disease images.