Segmentation of retinal blood vessels using gaussian mixture models and expectation maximisation

Djibril Kaba*, Ana G. Salazar-Gonzalez, Yongmin Li, Xiaohui Liu, Ahmed Serag

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present an automated method to segment blood vessels in fundus retinal images. The method could be used to support a non-intrusive diagnosis in modern ophthalmology for early detection of retinal diseases, treatment evaluation or clinical study. Our method combines the bias correction to correct the intensity inhomogeneity of the retinal image, and a matched filter to enhance the appearance of the blood vessels. The blood vessels are then extracted from the matched filter response image using the Expectation Maximisation algorithm. The method is tested on fundus retinal images of STARE dataset and the experimental results are compared with some recently published methods of retinal blood vessels segmentation. The experimental results show that our method achieved the best overall performance and it is comparable to the performance of human experts.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages105-112
Number of pages8
Volume7798 LNCS
DOIs
Publication statusPublished - 10 Apr 2013
Event2nd International Conference on Health Information Science, HIS 2013 - London, United Kingdom
Duration: 25 Mar 201327 Mar 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7798 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Conference

Conference2nd International Conference on Health Information Science, HIS 2013
CountryUnited Kingdom
CityLondon
Period25/03/1327/03/13

Keywords

  • bias correction
  • expectation maximisation
  • matched filter
  • Retinal image
  • vessel segmentation

Fingerprint

Dive into the research topics of 'Segmentation of retinal blood vessels using gaussian mixture models and expectation maximisation'. Together they form a unique fingerprint.

Cite this