A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images
Accurate optic disc (OD) segmentation and fovea detection in retinal fundus images are crucial for diagnosis in ophthalmology. We propose a robust and broadly applicable algorithm for automated, robust, reliable and consistent fovea detection based on OD segmentation. The OD segmentation is performe...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2017-09-01
|
| Series: | Current Directions in Biomedical Engineering |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/cdbme-2017-0113 |
| _version_ | 1797848745784639488 |
|---|---|
| author | Rust Caterina Häger Stephanie Traulsen Nadine Modersitzki Jan |
| author_facet | Rust Caterina Häger Stephanie Traulsen Nadine Modersitzki Jan |
| author_sort | Rust Caterina |
| collection | DOAJ |
| description | Accurate optic disc (OD) segmentation and fovea detection in retinal fundus images are crucial for diagnosis in ophthalmology. We propose a robust and broadly applicable algorithm for automated, robust, reliable and consistent fovea detection based on OD segmentation. The OD segmentation is performed with morphological operations and Fuzzy C Means Clustering combined with iterative thresholding on a foreground segmentation. The fovea detection is based on a vessel segmentation via morphological operations and uses the resulting OD segmentation to determine multiple regions of interest. The fovea is determined from the largest, vessel-free candidate region. We have tested the novel method on a total of 190 images from three publicly available databases DRIONS, Drive and HRF. Compared to results of two human experts for DRIONS database, our OD segmentation yielded a dice coefficient of 0.83. Note that missing ground truth and expert variability is an issue. The new scheme achieved an overall success rate of 99.44% for OD detection and an overall success rate of 96.25% for fovea detection, which is superior to state-of-the-art approaches. |
| first_indexed | 2024-04-09T18:32:31Z |
| format | Article |
| id | doaj.art-f3cbe0409b6f42afb728e70e5e9118cb |
| institution | Directory Open Access Journal |
| issn | 2364-5504 |
| language | English |
| last_indexed | 2024-04-09T18:32:31Z |
| publishDate | 2017-09-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Current Directions in Biomedical Engineering |
| spelling | doaj.art-f3cbe0409b6f42afb728e70e5e9118cb2023-04-11T17:07:14ZengDe GruyterCurrent Directions in Biomedical Engineering2364-55042017-09-013253353710.1515/cdbme-2017-0113cdbme-2017-0113A robust algorithm for optic disc segmentation and fovea detection in retinal fundus imagesRust Caterina0Häger Stephanie1Traulsen Nadine2Modersitzki Jan3Institute of Mathematics and Image Computing, University of Lübeck, Ratzeburger Allee 160, 23562 Lübeck, GermanyInstitute of Mathematics and Image Computing, University of Lübeck, Ratzeburger Allee 160, 23562 Lübeck, GermanyFraunhofer Institute for Medical Image Computing MEVIS, Maria-Goeppert-Straße 3, 23562 Lübeck, GermanyInstitute of Mathematics and Image Computing, University of Lübeck, Ratzeburger Allee 160, 23562 Lübeck, GermanyAccurate optic disc (OD) segmentation and fovea detection in retinal fundus images are crucial for diagnosis in ophthalmology. We propose a robust and broadly applicable algorithm for automated, robust, reliable and consistent fovea detection based on OD segmentation. The OD segmentation is performed with morphological operations and Fuzzy C Means Clustering combined with iterative thresholding on a foreground segmentation. The fovea detection is based on a vessel segmentation via morphological operations and uses the resulting OD segmentation to determine multiple regions of interest. The fovea is determined from the largest, vessel-free candidate region. We have tested the novel method on a total of 190 images from three publicly available databases DRIONS, Drive and HRF. Compared to results of two human experts for DRIONS database, our OD segmentation yielded a dice coefficient of 0.83. Note that missing ground truth and expert variability is an issue. The new scheme achieved an overall success rate of 99.44% for OD detection and an overall success rate of 96.25% for fovea detection, which is superior to state-of-the-art approaches.https://doi.org/10.1515/cdbme-2017-0113optic disc segmentationfovea detection |
| spellingShingle | Rust Caterina Häger Stephanie Traulsen Nadine Modersitzki Jan A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images Current Directions in Biomedical Engineering optic disc segmentation fovea detection |
| title | A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| title_full | A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| title_fullStr | A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| title_full_unstemmed | A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| title_short | A robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| title_sort | robust algorithm for optic disc segmentation and fovea detection in retinal fundus images |
| topic | optic disc segmentation fovea detection |
| url | https://doi.org/10.1515/cdbme-2017-0113 |
| work_keys_str_mv | AT rustcaterina arobustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT hagerstephanie arobustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT traulsennadine arobustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT modersitzkijan arobustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT rustcaterina robustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT hagerstephanie robustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT traulsennadine robustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages AT modersitzkijan robustalgorithmforopticdiscsegmentationandfoveadetectioninretinalfundusimages |