Using deep learning for robustness to parapapillary atrophy in optic disc segmentation

Ruchir Srivastava, Jun Cheng, Damon W.K. Wong, Jiang Liu

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

24 Citations (Scopus)

Abstract

Optic Disc (OD) segmentation from retinal fundus images is important for many applications such as detecting other optic structures and early detection of glaucoma. One of the major problems in segmenting OD is the presence of Para-papillary Atrophy (PPA) which in many cases looks similar to the OD. Researchers have used many different features to distinguish between PPA and OD, however each of these features has some limitation or the other. In this paper, we propose to use a deep neural network for OD segmentation which can learn features to distinguish PPA from OD. Using simple image intensity based features, the proposed method has the least mean overlapping error (9.7%) among the state-of-the-art works for OD segmentation in fundus images with PPA.

Original languageEnglish
Title of host publication2015 IEEE 12th International Symposium on Biomedical Imaging, ISBI 2015
PublisherIEEE Computer Society
Pages768-771
Number of pages4
ISBN (Electronic)9781479923748
DOIs
Publication statusPublished - 21 Jul 2015
Externally publishedYes
Event12th IEEE International Symposium on Biomedical Imaging, ISBI 2015 - Brooklyn, United States
Duration: 16 Apr 201519 Apr 2015

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2015-July
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Conference

Conference12th IEEE International Symposium on Biomedical Imaging, ISBI 2015
Country/TerritoryUnited States
CityBrooklyn
Period16/04/1519/04/15

Keywords

  • Optic Disc segmentation
  • deep learning
  • parapapillary atrophy
  • retinal fundus images

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Using deep learning for robustness to parapapillary atrophy in optic disc segmentation'. Together they form a unique fingerprint.

Cite this