Detection of Wet Age-related Macular Degeneration in OCT Images: A Case Study
[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ 2 ] Instytut Informatyki, Wydział Informatyki, Politechnika Poznańska | [ P ] employee
2018
chapter in monograph / paper
english
- wet aged-related macular degeneration
- OCT images
- gray level cooccurrence matrix
- feature extraction
- classification
EN Progress in medical imaging and computer vision has enabled us to rely on machines for the detection or diagnosis of many diseases, including eye-related problems. One of them is wet age-related macular degeneration (wet AMD) which is a type of age-related macular degeneration. Wet AMD causes the detachment of retinal pigment epithelium layer (RPE) – a condition referred to as pigment epithelium detachment (PED) – and also creates fluid fill region called choroidal neovascularization (CNV). In this paper we present a case study of detecting wet ADM in OCT images. We used a set of 51 images – 21 of sick eyes and 30 of healthy eyes. We employed feature extraction techniques to identify abnormalities in RPE layer (PED and CNV) along with the structural and textural properties of the RPE layer (gray level co-occurrence matrix, GLCM). Specifically, we considered four possible set of features and for each set we constructed k-NN, naive Bayes, support vector machine (SVM) and rule-based classifiers. The best classification performance was obtained for the features capturing the structural and textural properties of the RPE layer and for naive Bayes classifier (accuracy = 96.1%, sensitivity = 91.3%, specificity = 100.0%) and SVM classifier (accuracy = 94.1%, sensitivity = 100.0%, specificity = 93.7%). Our results confirm the usefulness of the features characterizing the RPE layer in the diagnosis of wet AMD.
43 - 51
20
WoS (15)