Share Email Print
cover

Journal of Biomedical Optics • Open Access

Fast retinal layer segmentation of spectral domain optical coherence tomography images
Author(s): Tianqiao Zhang; Zhangjun Song; Xiaogang Wang; Huimin Zheng; Fucang Jia; Jianhuang Wu; Guanglin Li; Qingmao Hu

Paper Abstract

An approach to segment macular layer thicknesses from spectral domain optical coherence tomography has been proposed. The main contribution is to decrease computational costs while maintaining high accuracy via exploring Kalman filtering, customized active contour, and curve smoothing. Validation on 21 normal volumes shows that 8 layer boundaries could be segmented within 5.8 s with an average layer boundary error <2.35 μm. It has been compared with state-of-the-art methods for both normal and age-related macular degeneration cases to yield similar or significantly better accuracy and is 37 times faster. The proposed method could be a potential tool to clinically quantify the retinal layer boundaries.

Paper Details

Date Published: 21 September 2015
PDF: 14 pages
J. Biomed. Opt. 20(9) 096014 doi: 10.1117/1.JBO.20.9.096014
Published in: Journal of Biomedical Optics Volume 20, Issue 9
Show Author Affiliations
Tianqiao Zhang, Shenzhen Institute of Advanced Technology (China)
Chinese Academy of Sciences (China)
Zhangjun Song, Shenzhen Institute of Advanced Technology (China)
Xiaogang Wang, Shanxi Eye Hospital (China)
Huimin Zheng, Shenzhen Institute of Advanced Technology (China)
Fucang Jia, Shenzhen Institute of Advanced Technology (China)
Jianhuang Wu, Shenzhen Institute of Advanced Technology (China)
Guanglin Li, Shenzhen Institute of Advanced Technology (China)
Qingmao Hu, Shenzhen Institute of Advanced Technology (China)


© SPIE. Terms of Use
Back to Top