Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6921703 | Computers in Biology and Medicine | 2014 | 9 Pages |
Abstract
This paper presents a methodology for glaucoma detection based on measuring displacements of blood vessels within the optic disc (vascular bundle) in human retinal images. The method consists of segmenting the region of the vascular bundle in an optic disc to set a reference point in the temporal side of the cup, determining the position of the centroids of the superior, inferior, and nasal vascular bundle segmented zones located within the segmented region, and calculating the displacement from normal position using the chessboard distance metric. The method was successful in 62 images out of 67, achieving 93.02% sensitivity, 91.66% specificity, and 91.34% global accuracy in pre-diagnosis.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
José Abel de la Fuente-Arriaga, Edgardo M. Felipe-Riverón, Eduardo Garduño-Calderón,