June 2013
Volume 54, Issue 15
Free
ARVO Annual Meeting Abstract  |   June 2013
Fully-automated multimodal co-registration of optical coherence tomography to colour fundus photography and fluorescein angiography
Author Affiliations & Notes
  • Pedro Guimaraes
    IBILI - Faculty of Medicine, University of Coimbra, Coimbra, Portugal
  • Pedro Rodrigues
    CNTM/AIBILI - Association for Innovation and Biomedical Research on Light and Image, Coimbra, Portugal
  • Conceicao Lobo
    CNTM/AIBILI - Association for Innovation and Biomedical Research on Light and Image, Coimbra, Portugal
  • Pedro Serranho
    IBILI - Faculty of Medicine, University of Coimbra, Coimbra, Portugal
    Mathematics Section, Department of Science and Technology, Open University, Lisbon, Portugal
  • Rui Bernardes
    IBILI - Faculty of Medicine, University of Coimbra, Coimbra, Portugal
    CNTM/AIBILI - Association for Innovation and Biomedical Research on Light and Image, Coimbra, Portugal
  • Footnotes
    Commercial Relationships Pedro Guimaraes, None; Pedro Rodrigues, None; Conceicao Lobo, None; Pedro Serranho, None; Rui Bernardes, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2013, Vol.54, 5539. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Pedro Guimaraes, Pedro Rodrigues, Conceicao Lobo, Pedro Serranho, Rui Bernardes; Fully-automated multimodal co-registration of optical coherence tomography to colour fundus photography and fluorescein angiography. Invest. Ophthalmol. Vis. Sci. 2013;54(15):5539.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To automatically co-register optical coherence tomography (OCT) to colour fundus photography (CFP), and fluorescein angiography (FA).

Methods: The multimodal co-registration of ocular fundus images of the human eye makes possible the integrated overview of the pathophysiology of retinal changes. While of the major interest, the number of images in the daily clinical practice and research, in addition to the burden of manually assisted image co-registration, prevents its widespread use. In this work we propose an automatic algorithm able to cope with retinal fundus images from high-definition OCT, CFP, and FA modalities. These are particularly challenging ones due to the clear distinct level of vascular network detail from one another. Images of 20 eyes from 13 patients diagnosed with type 2 diabetes mellitus that underwent high-definition OCT, CFP and FA, were collected and processed to compute the respective vascular tree. Typically, OCT fundus references translate into a poorly detailed vascular network, thus rendering difficult its co-registration to complementary imaging modalities. A recently developed method (by our research group) to compute the vascular network from high-definition OCT to the level of detail of the CFP was used in this work. Briefly each image is pre-processed to obtain crossovers and bifurcations from the vascular tree, and descriptors are generated from the triangulated branch-points. Inliers are automatically recruited in an iterative process followed by a stepwise regression and backward elimination. The process repeats until no further recruitment are possible, and the transformation matrix is computed from the set of established bifurcation/crossovers correspondences.

Results: To evaluate the obtained co-registration, a skeleton overlap metric was defined and computed based on the vascular tree skeleton (0 - no overlap, 1 - full overlap). From the co-registration of high-definition OCT and CFP images, an average skeleton overlap of 0.87 ± 0.053 (N=20) is reported. Similarly, the co-registration of high-definition OCT and FA renders an average Skeleton Overlap of 0.95 ± 0.026 (N=20).

Conclusions: The achieved level of co-registration, render this process an asset to the clinical daily practice and research.

Keywords: 549 image processing • 550 imaging/image analysis: clinical • 688 retina  
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×