Investigative Ophthalmology & Visual Science Cover Image for Volume 61, Issue 7
June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Comparison of Imaging Protocols for Screening Diabetic Retinopathy: 1-Field vs 2-Field
Author Affiliations & Notes
  • Nancy Barrett
    Fundus Photo Reading Center University of WI, Madison, Wisconsin, United States
  • Amitha Domalpally
    Fundus Photo Reading Center University of WI, Madison, Wisconsin, United States
  • Barbara A Blodi
    Fundus Photo Reading Center University of WI, Madison, Wisconsin, United States
  • Tyler Etheridge
    Fundus Photo Reading Center University of WI, Madison, Wisconsin, United States
  • James Reimers
    Fundus Photo Reading Center University of WI, Madison, Wisconsin, United States
  • Footnotes
    Commercial Relationships   Nancy Barrett, None; Amitha Domalpally, None; Barbara Blodi, None; Tyler Etheridge, None; James Reimers, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 3829. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Nancy Barrett, Amitha Domalpally, Barbara A Blodi, Tyler Etheridge, James Reimers; Comparison of Imaging Protocols for Screening Diabetic Retinopathy: 1-Field vs 2-Field. Invest. Ophthalmol. Vis. Sci. 2020;61(7):3829.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : The International Clinical Diabetic Retinopathy (ICDR) severity scale for determining referable and vision-threatening levels of Diabetic Retinopathy (DR) is accepted in both clinical and telemedicine screening settings. The imaging protocol for DR screening varies in the number of fields obtained, ranging from 1-3 fields. Most recently, validation of deep learning algorithms for DR screening has been performed using 2 field images. We evaluated DR grading using the ICDR scale in both single-field, non-stereo, macula-centered 45° images and in 2-field, non-stereo, 45° disc- and macula-centered imaging. We compared ICDR severity scale levels between the two imaging protocols.

Methods : DR severity was determined using ICDR 5-step scale. 60 subjects were graded first with only the single macula-centered 45° field available to the grader. After a 24-hour waiting period, the eyes were reassessed with both the single macula-centered field and an additional disc-centered 45° field available. All images were evaluated independently using double read.

Results : In the 60 sets of images evaluated no single-field images were deemed “Cannot Grade” and in the 2-field set of images 1 eye was deemed “Cannot Grade” due to photo quality. DR level was well-distributed with 21 (35.5%) “No DR”, 9 (15.3%) “Mild Non-Proliferative Diabetic Retinopathy (NPDR)”, 18 (30.5%) ”Moderate NPDR”, 3 (5.1%) “Moderate to Severe NPDR, and 8 (13.6%) “Proliferative DR” based on assessment of the 2-field image sets.
Comparing ICDR level for 1-field grading to 2-field grading for the same set of subjects, there was agreement in 79.6% (47), Kappa = 0.71. In eyes with disagreement between 1-field and 2-field grading, the DR level decreased in 5 (8.4%) eyes, and the DR level increased in 7 (11.9%) eyes. The 2-field grading demonstrated that there was 1 eye (1.7%) where DR level changed from non-referable DR to referable DR and 3 eyes (5.1%) where the DR level changed from non-vision-threatening DR to vision-threatening DR.

Conclusions : The study shows substantial agreement between ICDR levels determined by a single macula-centered 45° image compared to ICDR Level determined using 2 45° fields disc- and macula-centered imaging. Addition of a second field can potentially increase the detection of both referable and vision-threatening DR.

This is a 2020 ARVO Annual Meeting abstract.

×
×

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.

×