Segmentation of cyctoids macular edema in optical cohenrence tomography

Optical coherence tomography (OCT) using near infrared light is becoming one of the most important modalities for noninvasive assessment of ocular diseases. In many cases, processing OCT image is important for local retinal abnormalities detection. In this paper, we propose a method to identify and quantify cystoids macular edema regions in Spectral Domain OCT (SD-OCT) of the macula. A segmentation algorithm for SD-OCT was developed to detect cystoids within retina. Once the cystoids' borders are detected, a quantitative analysis of liquid in cystoids regions is performed to obtain surface area of fluid in the image. The performance of the proposed segmentation process for the cystoids macular edema was evaluated both qualitatively and quantitatively. Results showed a good agreement between quantitative evaluation and qualitative evaluation done by an ophthalmologist and proved that the process give an acceptable and accurate description of the cystoids regions.