Prognosis of Human Epidermal Malignancy: A Machine learning Approach
Sarvesh Nishad, Rohit Rohit, Kapil Sharma · 2021 3rd International Conference on Advances in Computing, Communication Control and Networking (ICAC3N) · 2021
Malignancy is tendency of medical condition to progressively become fatal and generally refers to the presence of cancerous cells with the ability to spread or to invade, and destroy tissues. Malignancy can start from any three layers of the epidermis, the probability of malignancy is upraised in those people where melanin pigment availability is genetically less. Increasing cases of malignant cancer have attracted researchers to find new approaches for their identification, and characterization. Various research studies have used Picture filtering and Computer’s vision that segregate malignancy into broadly seven category but Machine learning neural network models has a considerable advantage in clinical dermatology as these utilize VGGNet16, ResNet50, and CNN models. These models are applied for high-quality images with 128x128 and 256x256 resolution on HAM (HUMAN Against Machine) data sets having 10000 images. This approach focuses on the prediction of any classified skin malignancy taking patient image as a processing input. The commercial application will not only shorten the cost of the treatment but will also provide quick results for large population (around 3 million people) globally.