Segmentation of immunohistochemical staining of β-catenin expression of oral cancer using Gabor filter technique

In this paper, a system for diagnosis of oral cancer has been developed using Gabor filter technique. The microscopic images of immunohistochemical staining of β-catenin expression are segmented using a bank of Gabor filter tuned to different spatial frequencies and orientation in order to decompose the image into number of filtered image to obtain feature image. To produce segmented image Fuzzy C-means (FCM) , K-means (KM) and Thresholding (Th) techniques are applied. To evaluate the segmentation process, an objective quantitative evaluation function F for the segmented textured images using Gabor FCM, KM, Th (GFCM, GKM, GTh) is determined by varying number of regions. Based on the evaluation process we conclude that GKM algorithm is efficient technique to help the pathologist to evaluate the histological changes on microscopic images of oral cancer.