Skin Cancer Risks Estimation Using VGG19 Framework

The type of cancer which is most commonly found throughout the world is skin cancer. Before the cancer gets too serious, a premature detection of skin cancer is crucial for improving the patient’s health and reducing the rate of fatality (Mishra et al. in Rice yield estimation using deep learning [1]). In the research that we have conducted, we propose an approach for skin cancer detection using convolutional neural networks (CNNs). Our proposed system takes dermatoscopic skin lesions images as input and uses a CNN to classify the images into one of the seven classes which are provided in the HAM10000 dataset. We used a publicly available dataset containing 10,015 images to train and evaluate our CNN. Our proposed CNN achieves a high accuracy of 95.6%.