June 2023
Volume 64, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2023
Application of data-efficient generative techniques for Multi-Class Diabetic Retinopathy Classification
Author Affiliations & Notes
  • Melissa Du
    Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
    Singapore Eye Research Institute, Singapore, Singapore
  • Kabilan Elangovan
    Singapore Eye Research Institute, Singapore, Singapore
  • Gilbert Lim
    Singapore Eye Research Institute, Singapore, Singapore
    Duke-NUS Medical School, Singapore, Singapore
  • Daniel SW Ting
    Singapore Eye Research Institute, Singapore, Singapore
    Singapore National Eye Centre, Singapore, Singapore, Singapore
  • Footnotes
    Commercial Relationships   Melissa Du None; Kabilan Elangovan None; Gilbert Lim None; Daniel Ting None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2023, Vol.64, 2670. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Melissa Du, Kabilan Elangovan, Gilbert Lim, Daniel SW Ting; Application of data-efficient generative techniques for Multi-Class Diabetic Retinopathy Classification. Invest. Ophthalmol. Vis. Sci. 2023;64(8):2670.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : One of the main limiting factors of classification algorithms in the medical imaging domain is the lack of relevant class-specific data, especially for rare disease conditions, which often leads to suboptimal performance of classification models. In this study, we experimented the use of a Data Efficient Generative technique to synthesize class specific DR images. Generated images were used to improve the diversity of input datasets for the training of DR classification models, and the performance of these models were evaluated against the performance of models trained on existing DR datasets.

Methods : StyleGAN2-ADA was used as the generative technique to synthesize class-specific diabetic retinopathy images. The dataset consists of 5 classes: No DR (0), Mild DR (1), Moderate DR (2), Severe DR (3) and Proliferative DR (4). The study has 3 stages: 1) Balance out the EyePacs DR dataset with oversampling and use it to train a Densenet-121 classification model. 2) Train StyleGAN2-ADA to synthesize class specific DR images using conditional class generation. 3) Train and evaluate another Densenet-121 classification model with the original dataset supplemented by GAN-generated images while keeping the testing dataset and hyper parameters constant.

Results : The training of the StyleGAN2-ADA model converged to a Frechet Inception Distance (FID) of 3.465. The accuracy of the Densenet-121 model trained on the original balanced Kaggle dataset was 0.60. This increased to 0.76 when the dataset was supplemented with GAN-generated images. The f1-score for class 0 changed the least, and the f1-scores of other classes increased to match the performance of class 0. This indicates that the poor performance was originally the result of oversampling for classes with deficient data and that performance improves when each class has an equally diverse set of images.

Conclusions : The performance of models trained on DR datasets supplemented with GAN-generated images perform significantly better on classification tasks. This study validates the potential for StyleGAN2-ADA to generate realistic DR images and highlights the use of generative models as a data supplementation techniques as compared to conventional oversampling methods.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

Figure 1. Sample of images from original and generated datasets

Figure 1. Sample of images from original and generated datasets

 

Figure 2. Densenet-121 classification model performance trained on balanced Kaggle dataset vs GAN-supplemented dataset

Figure 2. Densenet-121 classification model performance trained on balanced Kaggle dataset vs GAN-supplemented dataset

×
×

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.

×