Particle swarm optimizer based epithelial layer segmentation in CIElab color space

The study of epithelial tissue is critical for the pathologist to determine its grade and the presence of various diseases. Proper segmentation of the epithelial layer plays a significant role in different disease detection like Oral Sub-mucous Fibrosis (OSF), multi-class grading of Oral Squamous Cell Carcinoma (OSCC), and Epithelial Biopsy Image Classifier, etc. In this study, our main aim is automatic segmentation of epithelial layer from pathology images with is very helpful for Computer-Aided Diagnosis (CAD) based system. Here author used particle swarm optimization (PSO) and k-means clustering (KM) methods in CIElab color space for oral epithelial layer segmentation and also used different color spaces and its component for comparative study. Experimental results showed that the CIElab color space gave the best result among all other color spaces. The segmented images are compared with ground truth images and PSO with CIElab color space outperformed all other methods with an accuracy of 98.43%.