Detecting Carcinoma Cells using Computer Vision

Cancer is a biggest and difficult medical condition in the world that is killing many lives every year. Even though many numbers of treatments available for the early stages of cancer, it does not show any symptoms until the very later stages even though a variety of diagnosis available for cancer the recent development in computer and computer technology is making a change in the early detection of cancer cells. This paper proposes the way to detect cancer in early stages in development using computer vision and machine learning processing techniques. First, we introduce the high dimensionality reduction on big data to reduce the dimensions of the data and then train a neural network to identify the location of the keratinization region in the image. Then segment the image of the keratinization region after that a second artificial neural network will analyse the features that are detected by the image segmentation to declare the image of a cancer is benign or malignant.