DocumentCode
3707983
Title
Disc segmentation and BMO-MRW measurement from SD-OCT image using graph search and tracing of three bench mark reference layers of retina
Author
Md. Akter Hussain;Alauddin Bhuiyan;Kotagiri Ramamohanarao
Author_Institution
Department of Computing and Information Systems, The University of Melbourne, Australia
fYear
2015
Firstpage
4087
Lastpage
4091
Abstract
Spectral Domain Optical Coherence Tomography (SD-OCT) imaging is a robust diagnostic tool for visualizing pathology in the retina for glaucoma and other diseases. Bruch´s Membrane Opening minimum rim width (BMO-MRW) of retina is a good measurable marker for distinguishing glaucoma and normal individuals. In this paper, we have proposed a novel automatic method to compute BMO-MRW from segmenting the Inner limiting membrane (ILM) and Bruch´s Membrane Opening (BMO). Our technique uses the location tracking of three bench mark reference (TBMR) layers of the retina that reduce search space with improving accuracy. This also allows locating the boundary of the optic disc. The accuracy was tested against manually detected disc boundary for 13 subjects. The precision, recall and F1 score shows more than 95% for disc area and unsigned mean error rate of BMO-MRW is 58.62 db 43.12 (Mean ± SD) μm.
Keywords
"Retina","Image edge detection","Optical imaging","Adaptive optics","Image segmentation","Pathology","Biomedical optical imaging"
Publisher
ieee
Conference_Titel
Image Processing (ICIP), 2015 IEEE International Conference on
Type
conf
DOI
10.1109/ICIP.2015.7351574
Filename
7351574
Link To Document