An efficient technique for skin cancer classification using deep learning

The development and spreading of abnormal cells in the skin of the human body assumed to be as skin cancer. There are different skin cancer types,. however, melanoma is the most critical one. The prediction of cancer at an early stage could help in better and improved treatment. In this research, a deep convolutional neural network-based technique to extract spatial information is developed as a method for skin cancer classification. For experimentation, the dataset includes real data that is collected from DHQ hospital Faisalabad, Pakistan. The classification results of the our proposed method are compared with state-of-the-art approaches while utilizing a reduced number of factors/feature vectors. The experiments exhibit that the classification accuracy is about 93.29% using our proposed method. The outcome of the experimental investigation shows that our method has higher accuracy in comparison with state-of-the-art techniques in literature.