Published Paper Details:

One of the most common malignancies in the world is Skin Cancer. The features of the illness can only be evaluated by a clinical assessment of skin lesions, but this process might take a long time and can be interpreted in many ways. Machine Learning and Deep Learning techniques have been developed to assist dermatologists in making an early and accurate diagnosis of Skin Cancer, which is critical for increasing patient survival rates. In this study, we give a comprehensive literature review of recent studies that have used deep learning to categorize skin lesions, with the goal of giving researchers who are just getting started in this field a good place to begin their investigations. The papers that thoroughly and clearly defined the methods used and presented the findings obtained were chosen after an extensive search of many internet databases using inclusion/exclusion criteria. Sixty-eight studies were chosen for further study, the vast majority of which use Deep Learning methods, in particular Artificial Neural Networks, for skin cancer detection and classification. Given the encouraging findings too far, it seems likely that these methods will be incorporated into clinical practice in the not-too-distant future. . The deep learning architectures such as Deep Convolutional Neural Networks are developed to utilize the Convolutional Neural Network. It is demonstrated that accuracy of deep learning model is improved to 98.8% from 92.3% of Convolutional Neural Network model.