Discrimination of retinal images containing bright lesions using sparse coded features and SVM

Abstract : Diabetic Retinopathy (DR) is a chronic progressive disease of the retinal mi-crovasculature which is among the major causes of vision loss in the world. The diagnosis of DR is based on the detection of retinal lesions such as mi-croaneurysms, exudates and drusen in retinal images acquired by a fundus camera. However, bright lesions such as exudates and drusen share similar appearances while being signs of different diseases. Therefore, discriminating between different types of lesions is of interest for improving screening performances. In this paper, we propose to use sparse coding techniques for retinal images classification. In particular, we are interested in discriminating between retinal images containing either exudates or drusen, and normal images free of lesions. Extensive experiments show that dictionary learning techniques can capture strong structures of retinal images and produce discriminant descriptors for classification. In particular, using a linear SVM with the obtained sparse coded features, the proposed method achieves superior performance as compared with the popular Bag-of-Visual-Word approach for image classification. Experiments with a dataset of 828 retinal images collected from various sources show that the proposed approach provides excellent discrimination results for normal, drusen and exudates images. It achieves a sensitivity and a specificity of 96.50% and 97.70% for the normal class; 99.10% and 100% for the drusen class; and 97.40% and 98.20% for the exudates class with a medium size dictionary of 100 atoms.
Type de document :
Article dans une revue
Computers in Biology and Medicine, Elsevier, 2015, 10.1016/j.compbiomed.2015.04.026. <10.1016/j.compbiomed.2015.04.026>
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01145892
Contributeur : Désiré Sidibé <>
Soumis le : lundi 27 avril 2015 - 11:57:02
Dernière modification le : mardi 28 avril 2015 - 01:02:44
Document(s) archivé(s) le : mercredi 19 avril 2017 - 07:15:27

Fichier

dr_paper_V7.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Désiré Sidibé, Ibrahim Sadek, Fabrice Mériaudeau. Discrimination of retinal images containing bright lesions using sparse coded features and SVM. Computers in Biology and Medicine, Elsevier, 2015, 10.1016/j.compbiomed.2015.04.026. <10.1016/j.compbiomed.2015.04.026>. <hal-01145892>

Partager

Métriques

Consultations de
la notice

139

Téléchargements du document

434