Cloud Based Solution for Skin Cancer Classification Using Machine Learning Models with Image Segmentation

According to the research 30,000 people are affected by skin cancer per year. Skin cancer is the unusual growth of skin cells. It occurs on the skin area which is exposed to sunlight. It is classified into two types - Melanoma or Benign. In our proposed work, we use deep learning concept in order to perform segmentation and classification of the lesions. We make use of the full resolution Convolutional Network (FrCN) to segment the skin cancer image. Then the segmented image is given as input to a deep residual network for classification.