MALIGNANT SKIN CANCER DETECTION USING CONVOLUTIONAL NEURAL NETWORKING

Abstract

The incidence of skin cancer is increasing by epidemic proportions. According to WHO, Skin Cancer is the world’s 6th most common cancer. It can be classified into Basal cell carcinoma, Squamous cell carcinoma and Melanoma among which Melanoma is more difficult to predict. By using this method, we can assist dermatologists to detect at an early stage as Computer Vision plays a vital role in diagnosis. In this paper, to detect skin cancer we are using machine learning-based algorithms. Traditionally classification algorithms are Convolutional neural networking which Consists of initialization, adding a convolutional layer, summing pooling layer, summing flattening layer, summing a dense layer, then compiling Convolutional neural networks and fitting the CNN model to a dataset. We used machine learning model architecture to determine if the skin images of the patients are harmful or harmless via using machine learning libraries provided in python. We have chosen this approach to be more precise and specific in recognizing about cancer and ultimately declining the mortality rate caused by it.