June 2017
Volume 58, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2017
Automated Analysis of Retinal Images for detection of Glaucoma based on Convolutional Neural Networks.
Author Affiliations & Notes
  • JUAN JOSE Jose GOMEZ
    Biomedical Image Technologies (BIT), Universidad Politécnica de Madrid, Madrid, Madrid, Spain
    Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Madrid, Spain
  • Clara I Sanchez
    Department of Radiology and Nuclear Medicine, Radboud University Medical Center, Nijmegen, Netherlands
  • Bart Liefers
    Department of Radiology and Nuclear Medicine, Radboud University Medical Center, Nijmegen, Netherlands
  • Freerk G Venhuizen
    Department of Radiology and Nuclear Medicine, Radboud University Medical Center, Nijmegen, Netherlands
  • Gianluca Fatti
    Parc de Salut Mar, Barcelona, Spain
  • Antonio Morilla-Grasa
    Institut Catala de Retina, Barcelona, Spain
  • Yasmin Cartagena
    Parc de Salut Mar, Barcelona, Spain
  • Alejandra Herranz Cabarcos
    Parc de Salut Mar, Barcelona, Spain
  • Andres Santos
    Biomedical Image Technologies (BIT), Universidad Politécnica de Madrid, Madrid, Madrid, Spain
    Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Madrid, Spain
  • María Jesús Ledesma-Carbayo
    Biomedical Image Technologies (BIT), Universidad Politécnica de Madrid, Madrid, Madrid, Spain
    Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Madrid, Spain
  • Alfonso Anton-Lopez
    Universitat Internacional de Catalunya, Barcelona, Spain
    Parc de Salut Mar, Barcelona, Spain
  • Footnotes
    Commercial Relationships   JUAN JOSE GOMEZ, None; Clara Sanchez, None; Bart Liefers, None; Freerk Venhuizen, None; Gianluca Fatti, None; Antonio Morilla-Grasa, None; Yasmin Cartagena, None; Alejandra Herranz Cabarcos, None; Andres Santos, None; María Ledesma-Carbayo, None; Alfonso Anton-Lopez, Aeire (C), Alcon-Novartis (F), Alcon-Novartis (C), BrudiLab (F), Santen (C)
  • Footnotes
    Support  Fondo de Investigaciones Sanitarias (FIS PI15/00412)
Investigative Ophthalmology & Visual Science June 2017, Vol.58, 706. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      JUAN JOSE Jose GOMEZ, Clara I Sanchez, Bart Liefers, Freerk G Venhuizen, Gianluca Fatti, Antonio Morilla-Grasa, Yasmin Cartagena, Alejandra Herranz Cabarcos, Andres Santos, María Jesús Ledesma-Carbayo, Alfonso Anton-Lopez; Automated Analysis of Retinal Images for detection of Glaucoma based on Convolutional Neural Networks.. Invest. Ophthalmol. Vis. Sci. 2017;58(8):706.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To assess the performance of deep learning architectures based on convolutional neural networks (CNN) for the diagnosis of glaucoma in screening campaigns using color fundus images.

Methods : Two independent data sets were used to develop and evaluate the proposed method. 1) 805 color fundus images with a field of view of 45 degrees, centered on the macula and including the optic disc (OD) from patients with age ranging from 55 to 86 years old included in a glaucoma detection campaign performed at Hospital Esperanza (Barcelona). Annotations were performed by eight observers having 8 to 26 years of clinical experience. 2) 101 images from the publicly available Drishti-GS retinal image dataset (http://cvit.iiit.ac.in/projects/mip/drishti-gs/mip-dataset2/Home.php). The total 906 images were further organized into a training, monitoring and test set according to a 60-20-20 split. The process to train and validate the CNN had 3 steps. 1) Preprocessing: the edges and the background were blurred to reduce the effect of the bright fringe and the border. Then patches centered at the OD of size 256x256x3 pixels were automatically segmented and scaled to values from 0 to 1. 2) Implementation: The architecture consisted of ten convolutional layers (32 filters 3x3 pixels size) followed by rectified linear units and spatial max-pooling. The network ends with a fully connected layer and a soft-max classifier which outputs a score from 0 to 1. The network was trained using stochastic gradient descent and a learning rate of 0.005. To avoid overfitting data augmentation was performed applying randomly translations, flipping and rotations during the training, and dropout with probability of 0.5. 3) Monitoring and evaluation: the training was completed after 50 epochs. To evaluate the classification capabilities of the algorithm, the area under the receiver operating characteristic curve (ROC) was calculated using the training set.

Results : An automatic classification algorithm based on CNN was developed. The present method achieved an area under the ROC of 0.894. The accuracy to identify healthy and glaucoma cases was 0.884 and 0.781 respectively, using a threshold of 0.5.

Conclusions : The good performance of the proposed CNN architecture suggests potential usefulness of these methods for an initial automatic classification of images in screening campaigns for glaucoma.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

×
×

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.

×