Spatio-temporal characterization of vessel segments applied to retinal angiographic images

  • Authors:

  • Y. Bouaoune;M. K. Assogba;J. C. Nunes;P. Bunel

  • Affiliations:

  • Laboratoire d'Etude et de Recherche en Instrumentation, Signaux et Systèmes L.E.R.I.S.S. Université Paris XII-Val de Marne, 61 Avenue du Général de Gaulle, 94010 Créteil C ...;Laboratoire d'Etude et de Recherche en Instrumentation, Signaux et Systèmes L.E.R.I.S.S. Université Paris XII-Val de Marne, 61 Avenue du Général de Gaulle, 94010 Créteil C ...;Laboratoire d'Etude et de Recherche en Instrumentation, Signaux et Systèmes L.E.R.I.S.S. Université Paris XII-Val de Marne, 61 Avenue du Général de Gaulle, 94010 Créteil C ...;Laboratoire d'Etude et de Recherche en Instrumentation, Signaux et Systèmes L.E.R.I.S.S. Université Paris XII-Val de Marne, 61 Avenue du Général de Gaulle, 94010 Créteil C ...

  • Venue:
  • Pattern Recognition Letters
  • Year:
  • 2003

Quantified Score

Hi-index 0.10

Visualization

Abstract

In this paper, we present a new approach of analysis and recognition of retinal vessel segments for the quantification of their shapes change due to alterations. This approach is based on a spatial followed by a temporal characterization of retinal fluorescein angiographic images. The spatial characteristics (coordinates and classification number i.e. numbering of segments) are associated to bifurcation points (bp), which are matched in temporal image pairs for vessel segments correspondence forming. The matching process uses spatial and temporal characteristics between the bp and their surrounded vessel segments by computing a coefficient of similarity measurement.The recognition of vessel segments will help ophthalmologists in quantifying changes in vessel shape parameters and detect the temporal evolution of some retinal pathologies as the Sickle Cell retinopathy in our case.