Image processing methods for computer-aided screening of diabetic retinopathy

Abstract : Diabetic retinopathy is the main cause of blindness among the middle-aged population. An early detection and adapted treatment considerably reduce the risk of sight loss. Medical authorities recommend an annual examination to diabetic patients. Several diabetic retinopathy screening programs have been deployed to enforce this recommendation. The aim of the TeleOphta project was to automatically detect normal examinations in a diabetic screening system, in order to reduce the burden on readers, and therefore serve more patients. This thesis proposes several methods to extract information linked to diabetic retinopathy lesions from color eye fundus images.The detection of exudates, microaneurysms and hemorrhages is discussed in detail. One of the main challenges of this work is to deal with clinical images, acquired by different types of eye fundus cameras, by different persons. Therefore the data base heterogeneity is high. New pre-processing methods, which perform not only normalization and denoising tasks, but also detect reflections and artifacts in the images, are proposed. Novel candidate segmentation methods based on mathematical morphology, and new textural and contextual features for lesion characterization, are proposed. A random forest algorithm is used to detect lesions among the candidates. The proposed methods make extensive use of new residue analysis methods.Moreover, three new publicly available retinal image databases, e-ophtha EX, e-ophtha MA and e-ophtha HM, respectively designed to develop and evaluate exudate, microaneurysms and hemorrhages detections methods, are proposed in this work. The images are extracted from the OPHDIAT telemedicine network for diabetic retinopathy screening. Manual annotations of the lesions are given in detail in these databases. The proposed algorithms are evaluated on these databases.The proposed methods have been integrated within the TeleOphta system, which is presented and evaluated on two large databases. Each patient record is classified into two categories: “To be referred” or “Normal”. The classification is based not only on the results of the presented methods, but also on image signatures provided by other partners, as well as on medical and acquisition-related information. The evaluation shows that the TeleOphta system can make about 2 times more patients benefit from the diagnosis service.
Document type :
Theses
Liste complète des métadonnées

Cited literature [121 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/tel-01083819
Contributor : Abes Star <>
Submitted on : Tuesday, November 18, 2014 - 10:11:04 AM
Last modification on : Monday, November 12, 2018 - 10:58:52 AM
Document(s) archivé(s) le : Thursday, February 19, 2015 - 10:20:27 AM

File

2014ENMP0024.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01083819, version 1

Citation

Xiwei Zhang. Image processing methods for computer-aided screening of diabetic retinopathy. Other. Ecole Nationale Supérieure des Mines de Paris, 2014. English. ⟨NNT : 2014ENMP0024⟩. ⟨tel-01083819⟩

Share

Metrics

Record views

1152

Files downloads

1743