Melanoma Skin Cancer Disease Detection Using Convolutional Neural Network

Abstract

Melanoma, a deadly skin cancer, is caused due to the development of a malignant tumor of melanocytes. Melanoma is the most harmful form of carcinoma, caused by pigment-making cells, called melanocytes. Melanoma is responsible for 75 percent of skin cancer deaths. The survival rate of patients over the past five years' analysis shows that only 15% of patients' survival has been chronically treated. The problem lies in the manual system. There are about six maximum colors in the mole, so the problematic thing is the image, there are several shades and it is difficult to recognize by human beings. So we have developed this problem by using flask and the Pre-Trained Deep-learning model which is trained and tested with a dataset from HAM10000. This paper aims to detect and treat skin cancer in patients at an early stage in order to reduce the mortality rate.