Classification of Oral Submucous Fibrosis using Convolutional Neural Network

The biology is disrupted for many reasons which are sometimes fathomable and sometimes not. The paramount factors can be genetic and variations acquired but both subsequently gives the catastrophic outcome in case of menacing disease such as cancer. The detection of it has been done and goes way back but newer technology is taking over every decade in order to make it more and more precise. As human intervention can lead to errors, automated detection can improve the accuracy. Therefore in this study, convolutional neural network (CNN) has been explored for detection of normal and different stages of oral submucous fibrosis from microscopic images of stained biopsy samples. Data pre-processing has been implemented before feeding the images into neural network and an overall accuracy of 99.4% has been achieved which shows the effectiveness of CNN for the same.