AUTOMATED THREE-DIMENSIONAL SEGMENTATION OF RETINAL OCT IMAGES

Narendra Kumar Patel, Tejus Thirumeni, Renu John, P. Vimal Prabhu

Abstract: Fourier Domain - Optical Coherence Tomography (FD-OCT) is a well-established imaging technique in ophthalmology. The segmentation of anatomical and pathological structures in ophthalmic images is essential for diagnosis and treatment of ocular diseases. However, FD-OCT gives large amount of data, which makes it infeasible to process manually. We propose an automatic segmentation algorithm to detect intra-retinal layers based on intensity and weighted gradient methods. The algorithm is tested upon the images of healthy volunteers obtained from Cirrus HD-OCT system.

Keywords: Optical Coherence Tomography, image segmentation, retinal layers, Canny edge detection, axial gradient

DOI: https://doi.org/10.15623/ijret.2015.0424010