A WEB APPLICATION FOR SKIN CANCER DETECTION USING DEEP LEARNING

Abstract

Skin cancer is one of the most serious types of cancer, and the rate of deaths is on the rise due to a lack of awareness about the symptoms and how to avoid them. As a result, early detection at an early stage is critical to prevent cancer from spreading skin cancer. Skin cancer classification using dermoscopic pictures is a popular application for deep learning-based classifiers. However, classifier evaluation is frequently limited to holdout data, disguising typical flaws such as confounding factor susceptibility. The problem we have taken is a binary classification problem. We have created a web application to diagnose skin cancer in this study that will accept photographs and other types. The backend is trained with deep learning models and check for skin image or others. It has finally classified the skin picture as benign or cancerous with 100% confidence, and the study took into account additional metrics like precision, recall, and F1-score metrics.