A new real-time retinal tracking system for image-guided laser treatment

• Published in: IEEE transactions on biomedical engineering 2002-09, Vol.49 (9), p.1059-1067
• Main Authors: Solouma, N.H., Youssef, A.-B.M., Badr, Y.A., Kadah, Y.M.
• Format: Article
• Language: English
• Subjects: Algorithms; Biological and medical sciences; Cameras; Choroidal Neovascularization - diagnosis; Choroidal Neovascularization - etiology; Choroidal Neovascularization - surgery; Chronic Disease; Coloring Agents; Computer interfaces; Computer Simulation; Diabetic Retinopathy - complications; Eye Movements - physiology; False Positive Reactions; Filters; Humans; Image Enhancement - methods; Image segmentation; Indocyanine Green; Laser Coagulation - instrumentation; Laser Coagulation - methods; Laser modes; Laser theory; Lighting; Macular degeneration; Medical sciences; Merging; Microscopy, Video - methods; Movement; Ophthalmoscopy - methods; Pattern Recognition, Automated; Real time systems; Retina; Retinal Neovascularization - diagnosis; Retinal Neovascularization - etiology; Retinal Neovascularization - surgery; Retinal Vessels - anatomy & histology; Retinal Vessels - physiopathology
• ISSN: 0018-9294; 1558-2531
• DOI: 10.1109/TBME.2002.802059

A new system is proposed for tracking sensitive areas in the retina for computer-assisted laser treatment of choroidal neovascularization (CNV). The system consists of a fundus camera using red-free illumination mode interfaced to a computer that allows real-time capturing of video input. The first image acquired is used as the reference image and utilized by the treatment physician for treatment planning. A grid of seed contours over the whole image is initiated and allowed to deform by splitting and/or merging according to preset criteria until the whole vessel tree is demarcated. Then, the image is filtered using a one-dimensional Gaussian filter in two perpendicular directions to extract the core areas of such vessels. Faster segmentation can be obtained for subsequent images by automatic registration to compensate for eye movement and saccades. An efficient registration technique is developed whereby some landmarks are detected in the reference frame then tracked in the subsequent frames. Using the relation between these two sets of corresponding points, an optimal transformation can be obtained. The implementation details of proposed strategy are presented and the obtained results indicate that it is suitable for real-time location determination and tracking of treatment positions.