May 2006
Volume 47, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2006
Assessment of Accuracy and Repeatability of Semi–Automated Analysis of Retinal Vascular Geometry in Normal and Diabetic Subjects
Author Affiliations & Notes
  • D.H. Steel
    Ophthalmology, Sunderland Eye Infirmary, Sunderland, United Kingdom
  • M.S. Habib
    Ophthalmology, Sunderland Eye Infirmary, Sunderland, United Kingdom
  • B. AL–Diri
    Computing and Informatics Department, Lincoln University, Lincoln, United Kingdom
  • J. Lowell
    Computing Science, Durham University, Durham, United Kingdom
  • A. Hunter
    Computing and Informatics Department, Lincoln University, Lincoln, United Kingdom
  • Footnotes
    Commercial Relationships  D.H. Steel, None; M.S. Habib, None; B. AL–Diri, None; J. Lowell, None; A. Hunter, None.
  • Footnotes
    Support  City Hospitals Sunderland
Investigative Ophthalmology & Visual Science May 2006, Vol.47, 2792. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      D.H. Steel, M.S. Habib, B. AL–Diri, J. Lowell, A. Hunter; Assessment of Accuracy and Repeatability of Semi–Automated Analysis of Retinal Vascular Geometry in Normal and Diabetic Subjects . Invest. Ophthalmol. Vis. Sci. 2006;47(13):2792.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : Evaluation of repeatability and short term varaibility of semi–automated measurements of retinal vascular geometry (RVG), using novel computer algorithms in normal and diabetic subjects.

Methods: : Digital fundus images of 30 diabetics and 30 normal subjects were evaluated. RVG measurements included bifurcation angles and junction exponents. Intra and inter–observer repeatability were assessed. Short–term individual variability of measurements was analyzed by repeat photography on the same day and at one week.

Results: : In 960 images, Intra and Inter–observer repeatability were 5.1° and 5.8° respectively. Short–term individual variability was 5.4° in normal and diabetic subjects. (Kappa=0.74)(95% CI=0.65–0.83)

Conclusions: : This novel computer algorithm for semi–automated analysis of retinal vascular geometrical measurements is reliable and repeatable. Bifurcation geometry does not vary significantly in the short term. It can thus be used to detect and monitor changes in vascular measurements in future comparative studies

Keywords: imaging/image analysis: non-clinical • retina • imaging/image analysis: clinical 
×
×

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.

×