Skin Lesions Diagnosis Using ML and DL Classification Models

Abstract

Skin cancer has been steadily growing for decades and is now the most prevalent form of cancer in humans. Recently, there has been a remarkable increase in the incidence of skin cancer. This presents a challenge as skin cancer lesions come in intricate and varied shapes and textures, making diagnosis difficult even for experts. To tackle this problem, the main objective of this paper is to utilize image classification techniques to diagnose skin cancer. Artificial neural networks, particularly convolutional networks, have shown great success in this field to distinguish between malignant and benign tumors. By leveraging the strengths of both Support Vector Machine and Convolutional Neural Network, we strive to improve the overall performance of skin cancer diagnosis. For experimentation, we used the ISIC dataset, which contains a collection of photos for melanoma skin cancer.