May 2006
Volume 47, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2006
Automatic Quantitative Analysis of Corneal Neovascularization in Rabbit Eyes Infected With Herpes Simplex Virus Type 1 (HSV–1) by Computer Software Program
Author Affiliations & Notes
  • Y.–Y. Shieh
    University of California, Irvine, Orange, CA
    Radiological Sciences,
  • G.–C. Perng
    University of California, Irvine, Orange, CA
    Ophthalmology,
  • Footnotes
    Commercial Relationships  Y. Shieh, None; G. Perng, None.
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science May 2006, Vol.47, 1646. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Y.–Y. Shieh, G.–C. Perng; Automatic Quantitative Analysis of Corneal Neovascularization in Rabbit Eyes Infected With Herpes Simplex Virus Type 1 (HSV–1) by Computer Software Program . Invest. Ophthalmol. Vis. Sci. 2006;47(13):1646.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : Corneal neovascularization (CNV), abnormal formation of blood vessels in the cornea, is a major vision threat, and is the leading cause of blindness. Digital ocular images are frequently captured from HSV infected individuals or animal eyes to evaluate the severity of CNV. Traditionally the measurement mainly relies on the subjective judgment of the researchers rather than objective measurements. The measurements are further complicated by the variations in object orientation, the camera distance, and the zoom factors on different images. As a result, it is difficult to compare results generated by different researchers. Thus consistent, accurate, and repeatable measurement technique is needed. We propose an automatic CNV evaluation software program based on image segmentation and analysis techniques to reduce the human variation in measurement by the same researcher at different times as well as that among different researchers.

Methods: : The prototype software is designed to analyze the digital image of the ocular surface captured from HSV infected animal eyes. Two parameters were used to measure the severity of CNV: the length of individual blood vessel visible in the cornea, and the percentage of neovascularization in the plexus of the cornea. The digital image noise was reduced by a signal adaptive order–statistic filtering algorithm and pixels indicative of blood vessel growing were separated from background by image segmentation through thresholding. An image thinning technique was then used to reduce each individual blood vessel to its skeleton before its length was measured. The percentage of CNV was calculated by dividing the vessel growing area consisting of pixels indicative of blood vessel growth over the corneal area. A set of 15 images taken from rabbit eyes infected with HSV–1 were used to evaluate the performance of the program.

Results: : Our results demonstrated the feasibility of automatic measurement of the severity of CNV by calculating the length of individual blood vessel and total CNV–manifested area in infected cornea.

Conclusions: : The designed computer software program provides a consistent way of measuring the length of individual blood vessel and the total CNV–manifested areas in infected cornea. Such quantitative characterization of CNV is not only providing much more accurate measurements than the subjective judgment by human evaluation, but free from subjective variations among researchers.

Keywords: imaging/image analysis: non-clinical • image processing • cornea: epithelium 
×
×

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.

×