Skin Cancer Classification using ConvNET

“Face is the Index of Mind” is what they say, and so is the case with “Skin is the Index of Body”. No matter what the complexion of the person is, no matter what region of the globe they belong to, humans have always had skin, and with that comes the issue of skin imperfections and skin diseases. In this generation where even a small blemish can lead to a lot of insecurities in a person, these are to be answered with a lot of precision and trust. This model is our idea of addressing the issue, when the subject being dealt here is Skin Cancer, the stakes are even higher and the chances to go wrong are extremely thin. Hence came the novel idea of developing this model. Since skin cancer is a serious issue that affects people of all skin types and regions. Our model aims to provide answers to major skin cancer types at the user’s fingertips. We have developed a machine learning model that uses a basic Convolutional Neural Network (CNN) to classify skin cancer images with high precision. The CNN model works by picking up patterns and details from images very effectively, just like how a human mind works. Our model has an accuracy of 71.03% in detecting skin cancer types. Our model is highly efficient and useful, and it stands tall on the pillars of pattern recognition and detail extraction from images.