June 2022
Volume 63, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2022
Ensemble learning for AMD prediction using retina OCT scans
Author Affiliations & Notes
  • Mousa Moradi
    Department of Biomedical Engineering, University of Massachusetts Amherst, Amherst, Massachusetts, United States
  • Tianxiao Huan
    Department of Ophthalmology & Visual Sciences, University of Massachusetts Chan Medical School, Worcester, Massachusetts, United States
  • Yu Chen
    Department of Biomedical Engineering, University of Massachusetts Amherst, Amherst, Massachusetts, United States
    University of Massachusetts Amherst Institute for Applied Life Sciences, Amherst, Massachusetts, United States
  • Xian Du
    Department of Mechanical and Industrial Engineering, University of Massachusetts Amherst, Amherst, Massachusetts, United States
    University of Massachusetts Amherst Institute for Applied Life Sciences, Amherst, Massachusetts, United States
  • Johanna Seddon
    Department of Ophthalmology & Visual Sciences, University of Massachusetts Chan Medical School, Worcester, Massachusetts, United States
  • Footnotes
    Commercial Relationships   mousa moradi None; Tianxiao Huan None; Yu Chen None; Xian Du None; Johanna Seddon J. Seddon, Laboratories THEA (C), Gemini Therapeutics, Inc and Apellis (I), Code C (Consultant/Contractor)
  • Footnotes
    Support  NIH R01-EY011309, R01- EY028602, American Macular Degeneration Foundation, Northampton, MA; The Macular Degeneration Center of Excellence, University of Massachusetts Chan Medical School, Department of Ophthalmology and Visual Sciences, Worcester, MA (JS). Worcester Foundation, Worcester, MA; Pilot Project Program grant, Center for Clinical and Translational Science, University of Massachusetts Chan Medical School (TH).
Investigative Ophthalmology & Visual Science June 2022, Vol.63, 732 – F0460. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mousa Moradi, Tianxiao Huan, Yu Chen, Xian Du, Johanna Seddon; Ensemble learning for AMD prediction using retina OCT scans. Invest. Ophthalmol. Vis. Sci. 2022;63(7):732 – F0460.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Age-related macular degeneration (AMD) is a progressive retinal disease and a common cause of blindness in people over age 50. AMD pathology can be detected by optical coherence tomography (OCT) and categorized into advanced and non-advanced AMD based on presence of geographic atrophy and/or neovascularization. Here we propose a stack-based ensemble deep learning method and demonstrate that it can improve non-advanced and advanced AMD detection using retina OCT scans.

Methods : Images from 150 subjects and 278 eyes diagnosed as non-advanced or advanced AMD obtained by Zeiss Cirrus OCT (Carl Zeiss Meditec, Inc., Dublin, CA, USA) were used in this study. Low-quality images were excluded before processing. The cleaned dataset had 767 images labeled as non-advanced and 663 images labeled as advanced AMD. The quality of retina images improved by contrast limited adaptive histogram equalization. Each image was divided into 4 patches without overlap between each other. Two center patches near the fovea were used, resulting in 2860 patch images in total. The processed images were split into 80% for training, 15% for validation, and 5% for testing. Base learners include 3 custom sequential models with 15, 23, and 25 hidden layers trained by stochastic gradient descent (SGD), rmsprop, and Adam optimizers, respectively. The weights calculated from each base model were used as an input feature for a meta model. Figure 1 shows the block diagram of the proposed model.

Results : Classification results for the base and ensemble models are shown in Table 1. Model 3 trained by Adam achieved slightly higher accuracy (87%) than model 1 (85%) and model 2 (84%) using SGD and rmsprop optimizers. By stacking base learners, the ensemble model can improve the accuracy, specificity, sensitivity up to 91%, 92%, 90.9%, respectively, on the test set.

Conclusions : Stack-based ensemble deep learning can improve the detection of non-advanced and advanced AMD.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.

 

Flowchart of the proposed ensemble model. The output weights from 3 convolutional neural networks (CNNs) are used as an input for meta model and the stack model of three base models (Model= [model1, model2, model3]) is trained on the same dataset. Hyperparameters for all models: 100 epochs, 32 batch size, and 256*256 input channel size.

Flowchart of the proposed ensemble model. The output weights from 3 convolutional neural networks (CNNs) are used as an input for meta model and the stack model of three base models (Model= [model1, model2, model3]) is trained on the same dataset. Hyperparameters for all models: 100 epochs, 32 batch size, and 256*256 input channel size.

 

Classification results for the base and ensemble models. F_score is harmonic mean of precision and sensitivity and is defined as:2×(precision×sensitivity)/(precision+sensitivity).

Classification results for the base and ensemble models. F_score is harmonic mean of precision and sensitivity and is defined as:2×(precision×sensitivity)/(precision+sensitivity).

×
×

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.

×