Automated Feature Selection for Pathogen Yeast Cryptococcus Neoformans

Due to large storage of images, it is highly requested to analyze images in a fast and efficient way. Data mining and pattern recognition methods have been widely used to understand the image knowledge deeply inside. Feature selection and extraction is the preprocessing step of data mining. Our approach to mine from Images, deals mainly with identification and extraction of unique features for analysing the pathogen conditions of yeast Cryptococcus Neoformans. Our automated model can determine which features can be used to identify variance pathogen condition. Different methods for extraction have been tried. Features extracted and techniques used are evaluated using the new test set images. Experimental results show that the features extracted by our automated data driven model are sufficient to identify the patterns from the images.