Investigative Ophthalmology & Visual Science Cover Image for Volume 65, Issue 7
June 2024
Volume 65, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2024
Prediction of Long-term Anatomic Treatment Outcomes for Diabetic Macular Edema using a Generative Adversarial Network
Author Affiliations & Notes
  • Jiwon Baek
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
    Ophthalmolgy, The Catholic University of Korea College of Medicine, Seoul, Seoul, Korea (the Republic of)
  • Ye He
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
  • Mehdi Emamverdi
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
  • Alireza Mahmoudi
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
  • Muneeswar Gupta Nittala
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
  • Giulia Corradetti
    DIRRL, Doheny Eye Institute, Los Angeles, California, United States
  • Michael S Ip
    Doheny Eye Institute Doheny Image Reading Center, Los Angeles, California, United States
  • SriniVas R Sadda
    Doheny Eye Institute Doheny Image Reading Center, Los Angeles, California, United States
  • Footnotes
    Commercial Relationships   Jiwon Baek None; Ye He None; Mehdi Emamverdi None; Alireza Mahmoudi None; Muneeswar Gupta Nittala None; Giulia Corradetti None; Michael Ip None; SriniVas Sadda 4DMT, Abbvie, Alexion, Allergan Inc., Alnylam Pharmaceuticals, Amgen Inc., Apellis Pharmaceuticals, Inc., Astellas, Bayer Healthcare Pharmaceuticals, Biogen MA Inc., Boehringer Ingelheim, Carl Zeiss Meditec, Catalyst Pharmaceuticals Inc., Centervue Inc., GENENTECH, Gyroscope Therapeutics, Heidelberg Engineering, Hoffman La Roche, Ltd., Iveric Bio, Janssen Pharmaceuticals Inc., Nanoscope, Notal Vision Inc., Novartis Pharma AG, Optos Inc., Oxurion/Thrombogenics, Oyster Point Pharma, Regeneron Pharmaceuticals Inc., Samsung Bioepis, Topcon Medical Systems Inc. , Code C (Consultant/Contractor), Carl Zeiss Meditec, Heidelberg Engineering, Optos Inc., Nidek, Topcon, Centervue, Code F (Financial Support), Carl Zeiss Meditec, Heidelberg Engineering, Nidek Incorporated, Novartis Pharma AG, Topcon Medical Systems Inc., Code R (Recipient)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2024, Vol.65, 2834. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jiwon Baek, Ye He, Mehdi Emamverdi, Alireza Mahmoudi, Muneeswar Gupta Nittala, Giulia Corradetti, Michael S Ip, SriniVas R Sadda; Prediction of Long-term Anatomic Treatment Outcomes for Diabetic Macular Edema using a Generative Adversarial Network. Invest. Ophthalmol. Vis. Sci. 2024;65(7):2834.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Long-term prediction for diabetic macular edema (DME) is important and this can be achieved using generative adversarial networks (GANs) with data from a randomized controlled trial. We evaluated post-treatment optical coherence tomography (OCT) images generated by GANs trained on data from a randomized controlled trial of anti-vascular endothelial growth factor (VEGF) treatment (q4w x 12) for DME (CRTH258B2305, KINGFISHER).

Methods : Three-hundred twenty-seven DME eyes of 327 patients with Spectralis OCT images were included in this analysis. OCT B-scan images through the foveal center at week-0, -4, -12, and -52, fundus photography, and retinal thickness maps from each patient were collected. Datasets were divided into training and validation (n = 297), and test (n = 30) sets. Input images for each model comprised either baseline B-scan alone or in combination with others. Predictive post-treatment OCT B-scan images were generated using GAN models and compared with real week-52 images.

Results : For 30 test images for GAN models trained with baseline OCT B-scans, 28, 29, 15, 20, and 30 acceptable OCT images were generated by cycleGAN, unitGAN, Pix2Pix, Pix2PixHD, and RegGAN, respectively (P<0.001). In comparison with real week-52 images, generated images showed sensitivity, specificity, and positive predictive values (PPV) for residual fluid ranging 0.500-0.917, 0.778-0.944, and 0.667-0.917 and those for hard exudate (HE) ranging 0.545-0.818, 0.895-1.000, and 0.750-1.000. RegGAN exhibited the highest values. RegGAN trained with multi-input images showed improved performance with sensitivity, specificity, and PPV for fluid and HE (0.818-0.909, 0.947-1.000, and 0.900-1.000; 0.818-0.909, 0.947, and 0.900-0.909, respectively).

Conclusions : OCT images generated by GAN models could predict the presence of residual fluid and hard exudate after long-term treatment of DME using continuous anti-VEGF therapy. Implementation of this tool may help predict which eyes will remain refractory after long-term treatment, thereby facilitating the establishment of a proper treatment regimen for these eyes.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

×
×

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.

×