May 2008
Volume 49, Issue 13
ARVO Annual Meeting Abstract  |   May 2008
An Image-Based Method for Quantitative Determination of Corneal Staining in Dry Eye Patients
Author Affiliations & Notes
  • R. L. Ornberg
    Alcon Research Ltd, Fort Worth, Texas
    R & D Pharmacology Screening,
  • G. Williams
    Alcon Research Ltd, Fort Worth, Texas
    R & D Pharmacology Screening,
  • M. Senchyna
    Alcon Research Ltd, Fort Worth, Texas
    R & D Allergy and Dry Eye,
  • Footnotes
    Commercial Relationships  R.L. Ornberg, Alcon Research Ltd., E; G. Williams, Alcon Research Ltd., E; M. Senchyna, Alcon Research Ltd., E.
  • Footnotes
    Support  None.
Investigative Ophthalmology & Visual Science May 2008, Vol.49, 2809. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      R. L. Ornberg, G. Williams, M. Senchyna; An Image-Based Method for Quantitative Determination of Corneal Staining in Dry Eye Patients. Invest. Ophthalmol. Vis. Sci. 2008;49(13):2809. doi:

      Download citation file:

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

  • Supplements

Purpose: : Corneal staining with fluorescein is an accepted diagnostic of corneal surface health used in the assessment of dry eye. However, procedures for quantifying fluorescein staining are limited to visual scoring with the aid of a slit lamp. To improve fluorescein stain analysis, we have developed a new imaged based method to quantify corneal surface staining in patients.

Methods: : 2% fluorescein solution applied for 3 min followed by two washes with saline. Fluorescein images with digital camera equipped with macro lens, a fiber optic ring light with flash. Incident light from the fiber ring light consist of a narrow blue (485 +/- 45 nm) wavelength ranged obtained with a bandpass filter designed for fluorescein excitation. Fluorescein emission is filtered with a 535 nm +/- 50 nm bandpass filter placed in the camera lens. Images are acquired with an f-stop ≥ 22 such that the entire ocular surface is in focus. Stained objects in the image are processed with a Fourier filter and segmented from background by thresholding routines. Stained objects are measured for size, intensity, and position relative to the central pupil.

Results: : Images produced by this system consist of fluorescein bound to surface defects and in the tear film. Artifacts such as the specular images of the ring light are removed by the spectral filters. The conjunctiva has a bright appearance that can exhibit surface defect staining. Fourier image processing designed to remove low frequency background results in flat gray images with bright objects coincident with regions of fluorescein stain in both the cornea and conjunctiva. Non-dry eye patients can exhibit a few small stain spots while dry eye patients typically have a band of punctate staining in the inferior cornea.

Conclusions: : The system easily identifies stained regions that are overlooked or unobserved with conventional cobalt blue slit lamp imaging. This system should be useful for quantitatively evaluating dry eye disease progression following therapy.

Keywords: cornea: tears/tear film/dry eye • imaging/image analysis: clinical • cornea: clinical science 

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.