June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Comparison of the performance of a Machine-to-Machine (M2M) deep learning algorithm versus human graders for glaucoma screening in teleretinal fundus images
Author Affiliations & Notes
  • Delaram Mirzania
    Duke University School of Medicine, Durham, North Carolina, United States
  • Atalie C Thompson
    Department of Ophthalmology, Duke University Hospital, Durham, North Carolina, United States
  • Eduardo Mariottoni
    Department of Ophthalmology, Duke University Hospital, Durham, North Carolina, United States
  • Leonardo Shigueoka
    Department of Ophthalmology, Duke University Hospital, Durham, North Carolina, United States
  • Kelly Muir
    Department of Ophthalmology, Duke University Hospital, Durham, North Carolina, United States
    Durham VA Medical Center, Durham, North Carolina, United States
  • Felipe Medeiros
    Department of Ophthalmology, Duke University Hospital, Durham, North Carolina, United States
  • Footnotes
    Commercial Relationships   Delaram Mirzania, None; Atalie Thompson, None; Eduardo Mariottoni, None; Leonardo Shigueoka, None; Kelly Muir, AbbVie (S), AbbVie (F), Arbutus Biopharma (S), Axovant (S), Blade (S), Cymabay (F), Dova (S), Dova (F), Gilead Sciences,, Inc. (S), Gilead Sciences, Inc. (F), Graybug vision (S), Merck Sharpe & Dohme (F), Merck Sharpe & Dohme (Merck & Co, USA) (S), NGM Biopharmaceuticals (F), Precision Bioscience (S), ProQR (S), Shionogi Pharma (S), Shire Pharmaceuticals (S), Taiwan J Pharmaceuticals (F); Felipe Medeiros, Aeri Pharmaceuticals (C), Allergan (C), Annexon (C), Biogen (C), Biozeus (C), Carl-Zeiss Meditec (F), Carl-Zeiss Meditec (C), Galimedix (C), Google (F), Heidelberg Engineering (F), IDx (C), NGoggle, Inc. (P), Novartis (C), Reichert (F), Reichert (C), Stealth Biotherapeutics (C)
  • Footnotes
    Support  NEI Grant EY029885 (FAM)
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 4545. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Delaram Mirzania, Atalie C Thompson, Eduardo Mariottoni, Leonardo Shigueoka, Kelly Muir, Felipe Medeiros; Comparison of the performance of a Machine-to-Machine (M2M) deep learning algorithm versus human graders for glaucoma screening in teleretinal fundus images. Invest. Ophthalmol. Vis. Sci. 2020;61(7):4545.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Application of deep learning (DL) algorithms may improve the efficacy of screening for glaucoma, especially if applied to fundus photos acquired by existing teleretinal screening programs. The purpose of this study was to apply and evaluate how a DL algorithm previously trained with optical coherence tomography (OCT) data to predict retinal nerve fiber layer (RNFL) thickness from photographs would perform compared to human graders in a real-world teleretinal screening program.

Methods : Cross-sectional study of 848 non-mydriatic fundus photos acquired in adult diabetics who underwent teleretinal screening at a community health center. Photographs were graded for glaucoma, no glaucoma, or deemed not gradable by two masked glaucoma specialists and disagreements were resolved by a third masked grader. A M2M DL algorithm that had been previously trained to predict OCT average retinal nerve fiber layer (RNFL) thickness from fundus photography was applied to the new set of photographs from teleretinal screening. The performance of the algorithm was compared to that of human graders. Generalized estimating equations were used to compare the predicted RNFL thickness between eyes that were classified with glaucoma or no glaucoma. The receiver operating characteristic (ROC) curve was plotted to illustrate the trade-off between sensitivity and specificity and the area under the ROC curve was calculated to summarize diagnostic accuracy.

Results : 4.1% of fundus photographs had glaucoma according to human graders, and 3.6% could not be graded. The mean predicted average RNFL thickness was 95.6 ± 10 mm for no glaucoma and 75.1 ± 8.9 mm for glaucoma (p < 0.001, GEE). The AUC for the DL algorithm’s ability to distinguish glaucoma from no glaucoma was 0.93 (95% CI: 0.90, 0.97) (Figure) if human graders were the reference standard.

Conclusions : Predictions of RNFL thickness from the M2M DL algorithm applied to fundus photographs were able to successfully discriminate glaucoma from non-glaucoma in the setting of an existing teleretinal screening program.

This is a 2020 ARVO Annual Meeting abstract.

 

Figure. Receiver operating characteristic curve to discriminate glaucoma from no-glaucoma for the machine-to-machine deep learning algorithm.

Figure. Receiver operating characteristic curve to discriminate glaucoma from no-glaucoma for the machine-to-machine deep learning algorithm.

×
×

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.

×