Search

Retinal veins act as an important part of the diagnosis and management of vision problems and a variety of eye diseases. To accurately segment the tiny vessels of the retina, the contrast is often very low, making the task challenging. Computer-based diagnosis of retinal vessels has become increasingly popular, with automated segmentation of retinal arteries becoming a key success using convolutional neural networks (CNNs). To identify the affected retinal vessels, two architectures, a manual architecture, and the standard UNet architecture were compared in this research to determine which performed with the highest accuracy. After the best architecture was determined, a model was created. This model was then used to produce a segmented image for the input image with the help of the CNN algorithm. With this technology, the diagnosis and treatment of a variety of eye diseases can be improved, as well as the identification of systemic diseases that can be seen through retinal changes. [ABSTRACT FROM AUTHOR]

Delineation of the optic disc boundary in retinal images is the first step towards the computation of cup-to-disc ratio, an important indicator of ophthalmic pathologies such as glaucoma. This paper proposes the combination of learning-based clustering trees with local mode filtering for the segmentation of the optic disc region in retinal images. The algorithm identifies candidate optic disc region by extracting and pooling low-level features at different clustering resolutions from the filtered region-of-interest in two color channels. Thereafter, we use learned geometric properties such as area, eccentricity and solidity to extract high-level features for the identification of connected components, which most likely belong to the optic disc region. The final stage pools and fully connects these connected components into a single segmented region. Performance evaluation on three publicly available datasets from IDRID, DRISHTI-GS and MESSIDOR demonstrate promising results that are comparable to state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]