A Comprehensive Literature Review of Skin Lesions Classification Techniques

An intelligent system that relies on image input to provide unbiased diagnoses of skin related diseases has become an indispensable screening tool. Healthcare employees have a hefty task due to the effort-intensive and slow process of seeing and manually diagnosing skin lesion conditions based on pictures, which is the typical technique. In order to reduce and alleviate this labor, a considerable deal of research has been done on machine learning and deep learning techniques. While early research focused more on normal machine learning approaches, more recent work has relied on deep learning techniques, mainly based on convolutional neural networks, which have shown amazing results. However, due to the complications with the skin lesions, some challenges remain open. Using deep learning approaches, this review paper offers a thorough study of the classification phases of skin lesions. Involving the classification methods utilized for skin lesion identification. This work examines and assesses the effectiveness of cutting-edge techniques that have been tried on a range of datasets including skin lesions. The research makes use of performance metrics for 7 distinct Convolutional Neural Network-based classification models.