July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Automated Development of Deep Learning Models to Diagnose Retinal Disease from Fundus and Optical Coherence Tomography Images
Author Affiliations & Notes
  • Pearse Andrew Keane
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
    NIHR Biomedical Research Center, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, London, United Kingdom
  • Livia Faes
    Retina, Eye Clinic, Cantonal Hospital Lucerne, Lucerne, Lucerne, Switzerland
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
  • Siegfried Wagner
    NIHR Biomedical Research Center, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, London, United Kingdom
  • Dun Jack Fu
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
  • Joseph R Ledsam
    DeepMind Technologies Ltd., United Kingdom
  • Reena Chopra
    NIHR Biomedical Research Center, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, London, United Kingdom
    DeepMind Technologies Ltd., United Kingdom
  • Christoph Kern
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
  • Gabriella Moraes
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
  • NIKOLAS PONTIKOS
    NIHR Biomedical Research Center, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, London, United Kingdom
  • Martin K Schmid
    Retina, Eye Clinic, Cantonal Hospital Lucerne, Lucerne, Lucerne, Switzerland
  • Lucas M Bachmann
    medignition Inc, Switzerland
  • Dawn A Sim
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
  • Konstantinos Balaskas
    Moorfields Eye Hospital NHS Foundation Trust, London, ENGLAND, United Kingdom
    NIHR Biomedical Research Center, Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology, London, London, United Kingdom
  • Footnotes
    Commercial Relationships   Pearse Keane, Allergan (R), Bayer (R), Bayer (S), Carl Zeiss Meditec (R), DeepMind (C), Haag-Streit (R), Heidelberg Engineering (R), Novartis (R), Novartis (S), Optos (C), Topcon (R); Livia Faes, None; Siegfried Wagner, None; Dun Jack Fu, None; Joseph Ledsam, Deep Mind Technologies Ltd. (E); Reena Chopra, DeepMind Technologies Ltd. (E); Christoph Kern, None; Gabriella Moraes, None; NIKOLAS PONTIKOS, None; Martin K Schmid, None; Lucas M Bachmann, None; Dawn Sim, Allergan (R), Allergan (S), Bayer (R), Bayer (S), Big Picture Eye Health (C), Haag-Streit (R), Novartis (R); Konstantinos Balaskas, Alimera (R), Allergan (R), Bayer (R), Bayer (F), Heidelberg engineering (R), Novartis (R), Novartis (F), TopCon (R)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 1453. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Pearse Andrew Keane, Livia Faes, Siegfried Wagner, Dun Jack Fu, Joseph R Ledsam, Reena Chopra, Christoph Kern, Gabriella Moraes, NIKOLAS PONTIKOS, Martin K Schmid, Lucas M Bachmann, Dawn A Sim, Konstantinos Balaskas; Automated Development of Deep Learning Models to Diagnose Retinal Disease from Fundus and Optical Coherence Tomography Images. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1453.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Deep learning (DL) has the potential to transform healthcare. But the highly specialized technical expertise that is necessary for its adoption is a considerable obstacle to its implementation. In this study, healthcare professionals (HCP) without any DL expertise sought to explore the feasibility of automated DL model development and to investigate their performance in diagnosing retinal disease from fundus and optical coherence tomography (OCT) images.

Methods : We used two open-source repositories, including retinal fundus (MESSIDOR) and OCT images (provided by Kermany et al.), to develop two separate DL models that predict retinal disease. Both datasets were fed into a neural architectural search framework that automatically curated the most suitable DL architecture and developed a DL model using a random split-sample validation. The DL models were trained to distinguish images showing diabetic retinopathy (DR) from healthy fundus (MESSIDOR), and neovascular age-related macular degeneration (nvAMD), diabetic macular oedema (DMO), drusen, and healthy OCT-images (Kermany et al.). Recall and precision were used to evaluate the diagnostic properties and the area under the curve (AUC) of plotted receiver operating characteristics curves were used to assess the discriminative performance.

Results : The MESSIDOR dataset involved 1187 images, with 153 images showing mild, 247 moderate, and 254 severe DR. The AUC was 0.87, best accuracy was reached at a cut-off value of 0.5 with a recall of 73.3% and a precision of 73.3%. The dataset provided by Kermany et al. involved 101,418 images of 5,761 patients. 31882 images depicted OCT changes related to nvAMD, 11165 to DMO, 8061 depicted drusen, and 50310 were normal. The AUC was 0.99, best accuracy was reached at a cut-off value of 0.5 with a recall of 97.3% and a precision of 97.7%.

Conclusions : In this preliminary study, automatically developed DL models showed high levels of diagnostic performance and discrimination for the diagnosis of common retinal diseases. This suggests that development of DL prediction models is feasible for (HCP) without any DL expertise. However, the translation of this technological success to meaningful clinical impact constitutes the next great challenge.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

×
×

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.

×