Retina analysis in optical coherence tomography images

Abstract

A new method of the OCT (Optical Coherence Tomography) image analysis is presented. It concerns eyeball examination in ophthalmology. Adaptive noise suppression algorithm, based on anisotropic diffusion, followed by multistep image analysis approach yield a segmentation and measurement of various layers within the retina structure. The proposed segmentation approach utilizes different image processing techniques including watershed transform, Fuzzy C-Means and HMinima Transform. Both, anatomical and pathological cases have been subjected to the analysis. The results have been evaluated by an expert.