28 February 2014 Segmentation method for in vivo meibomian gland OCT image
Author Affiliations +
Proceedings Volume 8930, Ophthalmic Technologies XXIV; 89300P (2014) https://doi.org/10.1117/12.2038886
Event: SPIE BiOS, 2014, San Francisco, California, United States
Abstract
We demonstrate segmentation of human MGs based on several image processing technic. 3D volumetric data of upper eyelid was acquired from real-time FD-OCT, and its acini area of MGs was segmented. Three dimensional volume informations of meibomian glands should be helpful to diagnose meibomian gland related disease. In order to reveal boundary between tarsal plate and acini, each B-scan images were obtained before averaged three times. Imaging area was 10x10mm and 700x1000x500 voxels. The acquisition time was 60ms for B-scan and 30sec for C-scan. The 3D data was flattened to remove curvature and axial vibration, and resized to reduce computational costs, and filtered to minimize speckles, and segmented. Marker based watershed transform was employed to segment each acini area of meibomian gland.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jun Geun Shin, Byeong Ha Lee, Tae Joong Eom, "Segmentation method for in vivo meibomian gland OCT image", Proc. SPIE 8930, Ophthalmic Technologies XXIV, 89300P (28 February 2014); doi: 10.1117/12.2038886; https://doi.org/10.1117/12.2038886
PROCEEDINGS
6 PAGES


SHARE
Back to Top