June 2020
Volume 61, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2020
Deep Learning for Predicting the Progression of Diabetic Retinopathy using Fundus Images
Author Affiliations & Notes
  • Ashish Bora
    Google Health, Google, Mountain View, California, United States
  • Boris Babenko
    Google Health, Google, Mountain View, California, United States
  • Sunny Virmani
    Verily Life Sciences LLC, California, United States
  • Jorge Cuadros
    EyePACS LLC, California, United States
  • Siva Balasubramanian
    Work done at Google Health via Advanced Clinical, Illinois, United States
  • Avinash V. Varadarajan
    Google Health, Google, Mountain View, California, United States
  • Footnotes
    Commercial Relationships   Ashish Bora, Google (E); Boris Babenko, Google (E); Sunny Virmani, Verily Life Sciences (E); Jorge Cuadros, EyePACS (E); Siva Balasubramanian, Google (C); Avinash Varadarajan, Google (E)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 1639. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ashish Bora, Boris Babenko, Sunny Virmani, Jorge Cuadros, Siva Balasubramanian, Avinash V. Varadarajan; Deep Learning for Predicting the Progression of Diabetic Retinopathy using Fundus Images. Invest. Ophthalmol. Vis. Sci. 2020;61(7):1639.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Patients with Diabetic Retinopathy (DR) can reduce chances of vision loss via regular screening and timely treatment. However, the current standard of care is based on stratifying patients into a small number of risk groups. More accurate quantification of progression risk could allow improved care via personalized disease management. In this work, we developed and validated a deep learning (DL) algorithm using primary field color fundus photographs (CFPs) to predict 6, 12, and 24 month progression to DR.

Methods : A DL algorithm was developed using a longitudinal dataset of 664,622 CFPs retrospectively collected across 367,146 visits from 574,860 eyes of 289,826 subjects with diabetes (mean age: 54 yrs, 60% women). The images were independently graded for DR (26% prevalence). The resultant algorithm was evaluated using an independent validation dataset with 166,661 CFPs across 91,942 visits from 144,003 eyes of 72,457 diabetic patients (mean age: 54 yrs, 61% women, 26% DR prevalence).

Results : For predicting the progression to DR, the algorithm had an area under the receiver operating characteristic curve (AUC) of 0.73 (95% confidence interval (CI), 0.68-0.78), 0.71 (95% CI, 0.69-0.74), 0.72 (95% CI, 0.71-0.74) at 6, 12, and 24 months respectively (Figure 1). Kaplan-Meier plots also show stratification of low and high risk groups for progression to DR (Figure 2).

Conclusions : In this study, our DL model predicted the future risk of progression to DR using primary field CFPs. Further research is necessary to determine the feasibility of applying this algorithm in the real world clinical setting for optimizing diagnosis and management of DR.

This is a 2020 ARVO Annual Meeting abstract.

 

Figure 1. ROC curve of the DL model for progression to DR in 6, 12, and 24 months. For each ROC curve, the legend shows the AUC along with 95% confidence intervals. The total number of eyes (N) and the number of eyes that progressed to DR (n) are also shown for each ROC curve.

Figure 1. ROC curve of the DL model for progression to DR in 6, 12, and 24 months. For each ROC curve, the legend shows the AUC along with 95% confidence intervals. The total number of eyes (N) and the number of eyes that progressed to DR (n) are also shown for each ROC curve.

 

Figure 2. Kaplan-Meier plots for progression to DR. Eyes with 24 month progression risk at the first visit greater (smaller) than predefined threshold were defined as the high (low) risk group. The shaded regions indicate 95% confidence intervals.

Figure 2. Kaplan-Meier plots for progression to DR. Eyes with 24 month progression risk at the first visit greater (smaller) than predefined threshold were defined as the high (low) risk group. The shaded regions indicate 95% confidence intervals.

×
×

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.

×