July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Non-Enhanced vs Software Enhanced Images of the Optic Nerve Head and Nerve Fiber Layer in tele-glaucoma screening
Author Affiliations & Notes
  • Subhashini Chandrasekaran
    Rutgers New Jersey Medical School, Newark, New Jersey, United States
  • Sumana Kommana
    Temple University, Pennsylvania, United States
  • Ben Szirth
    Rutgers New Jersey Medical School, Newark, New Jersey, United States
  • Albert S Khouri
    Rutgers New Jersey Medical School, Newark, New Jersey, United States
  • Footnotes
    Commercial Relationships   Subhashini Chandrasekaran, None; Sumana Kommana, None; Ben Szirth, None; Albert Khouri, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 5521. doi:https://doi.org/
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Subhashini Chandrasekaran, Sumana Kommana, Ben Szirth, Albert S Khouri; Non-Enhanced vs Software Enhanced Images of the Optic Nerve Head and Nerve Fiber Layer in tele-glaucoma screening. Invest. Ophthalmol. Vis. Sci. 2019;60(9):5521. doi: https://doi.org/.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Over 2.7 million individuals are affected by glaucoma in the USA, half of whom are not aware of it until vision complications manifest themselves. Community-based screenings rely on images captured by a non-mydriatic retinal camera taken through small pupils or cataracts that may obscure retinal findings. Ophthalmic imaging software was developed in 2016 to post-capture enhance images to offer a better view of the eye. The purpose of this prospective pilot study is to compare gradeability of original vs. Opacity Suppression (OSTM) software enhanced images.

Methods : We evaluated the benefits of original retinal images of participants with glaucoma captured with a Canon 45-degree field of view non-mydriatic retinal camera vs digitally automated software enhanced images using an OSTM filter (N=93; 170 fundi). All images were initially analyzed by 3 graders and a cloud based Artificial Intelligence software (AI) (Pegasus, UK) for overall image quality, optic nerve head (ONH) and nerve fiber layer (NFL) appearance. Images were then reanalyzed after being passed through the OSTM software to evaluate qualitatively each structure and findings. The analysis was done on a 1 to 5 grading system(Table 1). A grade of 1 was assigned for images of no clinical value and a grade of 5 was representative of a perfect image where the reader was able to evaluate all subtle findings.

Results : The main benefit of OSTM enhancement was for NFL appearance and retinal view improvement by 35%. NFL grade was higher in OSTM images (p=0.001) for all three graders. For NFL, the average grade for the readers was 2.81 for the original images and 3.17 for the OSTM images. There was no apparent benefit in OSTM enhanced images for ONH, due mostly to bleaching of rim color making it more challenging to evaluate the health of the ONH, and overall image grade (Table 2).

Conclusions : OSTM filter was able to provide an enhanced view of the NFL and retina in comparison to non-enhanced posterior pole images in subjects with clinically confirmed glaucoma. However, OSTM benefits did not extend to the ONH due to bleaching of the ONH. The OSTM filter could be helpful in glaucoma screening analysis to offer a better view of the NFL and retina. AI was not significantly affected by OSTM.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

Table 1: Classification of the grading system for analysis of images

Table 1: Classification of the grading system for analysis of images

 

Table 2: Summary of human readers and AI grading for original vs OSTM images

Table 2: Summary of human readers and AI grading for original vs OSTM images

×
×

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.

×