Suppr超能文献

通过双重配准和眼介质光学矫正从立体图像实现三维视盘可视化。

Three dimensional optic disc visualisation from stereo images via dual registration and ocular media optical correction.

作者信息

Xu J, Chutatape O, Zheng C, Kuan P C T

机构信息

Biomedical Engineering Research Centre, Nanyang Technological University, 50 Nanyang Drive, Research Techno Plaza, 6th Storey, XFrontiers Block, Singapore.

出版信息

Br J Ophthalmol. 2006 Feb;90(2):181-5. doi: 10.1136/bjo.2005.082313.

Abstract

BACKGROUND/AIMS: The three dimensional (3-D) visualisation of the optic disc in true colour will give essential meaning in clinical application. It is not only useful for clinicians in the evaluation of the condition of the optic disc, but it also simplifies the pathological diagnosis and disease progression monitoring. This paper describes a complete 3-D optic disc reconstruction method from a pair of stereo images by a series of robust procedures including camera calibration, image registration, depth recovery, and ocular media optical inclusion.

METHODS

Two registration techniques (correlation and feature based methods) are combined together to prune the uncertain matching points in order to improve the overall accuracy of registration. The ocular media within the eyeball are lump modelled as a single lens and integrated into the reconstruction process to obtain an accurate 3-D optic disc image.

CONCLUSION

The recovered 3-D optic disc images show good consistency and compatibility when compared with the results from Heidelberg retina tomography (HRT) under clinical validation, with an additional advantage of implementing a more economical and conventional mode of retinal image acquisition.

摘要

背景/目的:对视盘进行真彩色三维(3-D)可视化在临床应用中具有重要意义。它不仅有助于临床医生评估视盘状况,还能简化病理诊断和疾病进展监测。本文描述了一种通过一系列稳健程序从一对立体图像重建完整3-D视盘的方法,这些程序包括相机校准、图像配准、深度恢复和眼介质光学包含。

方法

将两种配准技术(相关法和基于特征的方法)结合起来,以去除不确定的匹配点,从而提高配准的整体准确性。眼球内的眼介质被整体建模为一个单透镜,并整合到重建过程中,以获得准确的3-D视盘图像。

结论

在临床验证中,与海德堡视网膜断层扫描(HRT)的结果相比,恢复的3-D视盘图像显示出良好的一致性和兼容性,并且具有采用更经济和传统的视网膜图像采集模式的额外优势。

相似文献

2
Auto-adjusted 3-D optic disk viewing from low-resolution stereo fundus image.
Comput Biol Med. 2006 Sep;36(9):921-40. doi: 10.1016/j.compbiomed.2005.05.001. Epub 2005 Jul 14.
3
3-d optic disk reconstruction via combined registration and inclusion of eye optical effects.
Conf Proc IEEE Eng Med Biol Soc. 2005;2005:3284-7. doi: 10.1109/IEMBS.2005.1617178.
4
3-D retinal surface inference: stereo or monocular fundus camera?
Annu Int Conf IEEE Eng Med Biol Soc. 2007;2007:896-9. doi: 10.1109/IEMBS.2007.4352435.
6
Digital stereo image analyzer for generating automated 3-D measures of optic disc deformation in glaucoma.
IEEE Trans Med Imaging. 2002 Oct;21(10):1244-53. doi: 10.1109/TMI.2002.806293.
7
Three-dimensional retinal imaging with high-speed ultrahigh-resolution optical coherence tomography.
Ophthalmology. 2005 Oct;112(10):1734-46. doi: 10.1016/j.ophtha.2005.05.023.

引用本文的文献

1
Automated volumetric evaluation of stereoscopic disc photography.
Opt Express. 2010 May 24;18(11):11347-59. doi: 10.1364/OE.18.011347.
2
Automated assessment of the optic nerve head on stereo disc photographs.
Invest Ophthalmol Vis Sci. 2008 Jun;49(6):2512-7. doi: 10.1167/iovs.07-1229. Epub 2008 Mar 7.

本文引用的文献

1
Automated detection of optic disk boundary by a new deformable model technique.
Conf Proc IEEE Eng Med Biol Soc. 2005;2005:6516-9. doi: 10.1109/IEMBS.2005.1615992.
2
Registration of bimodal retinal images - improving modifications.
Conf Proc IEEE Eng Med Biol Soc. 2004;2004:1695-8. doi: 10.1109/IEMBS.2004.1403510.
3
Auto-adjusted 3-D optic disk viewing from low-resolution stereo fundus image.
Comput Biol Med. 2006 Sep;36(9):921-40. doi: 10.1016/j.compbiomed.2005.05.001. Epub 2005 Jul 14.
5
Optical coherence tomography to detect and manage retinal disease and glaucoma.
Am J Ophthalmol. 2004 Jan;137(1):156-69. doi: 10.1016/s0002-9394(03)00792-x.
6
Digital stereo image analyzer for generating automated 3-D measures of optic disc deformation in glaucoma.
IEEE Trans Med Imaging. 2002 Oct;21(10):1244-53. doi: 10.1109/TMI.2002.806293.
7
Multiresolution image registration for two-dimensional gel electrophoresis.
Proteomics. 2001 Jul;1(7):856-70. doi: 10.1002/1615-9861(200107)1:7<856::AID-PROT856>3.0.CO;2-R.
8
A multimodal registration algorithm of eye fundus images using vessels detection and Hough transform.
IEEE Trans Med Imaging. 1999 May;18(5):419-28. doi: 10.1109/42.774169.
9
Registration of stereo and temporal images of the retina.
IEEE Trans Med Imaging. 1999 May;18(5):404-18. doi: 10.1109/42.774168.

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验