Automatic Retinal Vessel Segmentation

Diabetic Retinopathy is the most common cause of blindness in the working population of the western world and is very common among people who suffer from diabetes. Fortunately, during a clinical examination an ophthalmologist is able to determine the onset of the disease by taking certain features of the retinal vessels of the fundus into account. These features include the narrowing of vessels and their general structure. The clinical examination involves retinal imaging whereby a photo of the back of the inside of the eye is taken. The ophthalmologist then usually digitises the image by manually selecting what parts of the photo constitute vessels, so that certain statistics such as the thickness of the vessels can be calculated by a computer program. Labelling all the vessels however is a tedious and time consuming process. Our work focuses on using image processing techniques in order to develop a computer program that can automatically detect and segment blood vessels in these images, thereby saving the ophthalmologist considerable time. We make use of a fast level set method without solving partial differential equations to extract the contour of the vessels in the retina and overlay the contour over the original input image. This further aids the work of an ophthalmologist because it highlights the vessel structure.