Approach to retina optical coherence tomography image processing

Optical coherence tomography (OCT) is a recently developed imaging technology. By using Zeiss STRATUS OCT one could obtain clear tomography pictures of retina and macula lutea. The clinical use of image processing requires both medical knowledge and expertise in image processing techniques. This paper focused on processing of retina OCT image to design the automatic retina OCT image identification system that could help us evaluate retina, examine and clinically diagnose the fundus diseases. The motivation of our work is to extract the contour and highlight the feature area of the focus clearly and exactly. Generally it is related to image segmentation, enhancement and binarization etc. In the paper we try to reduce the image noise and make the symbolic area connected through color segmentation, low-pass filter and mathematical morphology algorithm etc., finally discern some common and different properties of postprocessing image compared with the real OCT image. Experiments were done on cystoid macular edema, macular hole and normal retina OCT image. The results show that the project raised is feasible and suitable for further image identification and classification according to ophthalmology criteria.