GMM clustering based segmentation and optic nervehead geometry detection from OCT nervehead images

Optical Coherence Tomography (OCT) is a non-invasive imaging technique which is widely used in the field of ophthalmology. Ophthalmology requires segmentation of retinal layers of OCT images for efficient diagnosis of diseases. A segmentation algorithm based on a kernel and clustering method can be used for the extraction of retinal boundary and choroid boundary of optic nervehead in OCT images. Precise segmentation is achieved by incorporating Gaussian mixture model (GMM) clustering into the kernel. The ratio of the area of the central portion of the optic nerve (the cup) to that of the complete nerve (the disk) is used for the diagnosis of glaucoma. As glaucoma progresses the cup to disc ratio increases. This Optic cup and disk end points can be detected by applying a curve partitioning method and a feature classification method on the extracted retinal and choroid boundary.