Training Deep Learning Models to Work on Multiple Devices by Cross-Domain Learning with No Additional Annotations.

Journal: Ophthalmology
Published Date:

Abstract

PURPOSE: To create an unsupervised cross-domain segmentation algorithm for segmenting intraretinal fluid and retinal layers on normal and pathologic macular OCT images from different manufacturers and camera devices.

Authors

  • Yue Wu
    Key Laboratory of Luminescence and Real-Time Analytical Chemistry (Ministry of Education), College of Pharmaceutical Sciences, Southwest University, Chongqing 400716, China.
  • Abraham Olvera-Barrios
    Medical Retina, Moorfields Eye Hospital NHS Foundation Trust, London, UK a.olvera@nhs.net.
  • Ryan Yanagihara
    Department of Ophthalmology, University of Washington, Seattle, Washington.
  • Timothy-Paul H Kung
    Department of Ophthalmology, University of Washington, Seattle, Washington.
  • Randy Lu
    From the Department of Ophthalmology, University of Washington, Seattle, Washington.
  • Irene Leung
    Moorfields Eye Hospital NHS Foundation Trust, London, United Kingdom.
  • Amit V Mishra
    Moorfields Eye Hospital NHS Foundation Trust, London, United Kingdom.
  • Hanan Nussinovitch
    Moorfields Eye Hospital NHS Foundation Trust, London, United Kingdom.
  • Gabriela Grimaldi
    Moorfields Eye Hospital NHS Foundation Trust, London, United Kingdom.
  • Marian Blazes
    From the Department of Ophthalmology, University of Washington, Seattle, Washington.
  • Cecilia S Lee
    Department of Ophthalmology, University of Washington, Seattle, Washington, USA.
  • Catherine Egan
    NIHR Biomedical Research Centre at Moorfields Eye Hospital and UCL Institute of Ophthalmology, London, UK.
  • Adnan Tufail
    London, United Kingdom. Electronic address: Adnan.Tufail@moorfields.nhs.uk.
  • Aaron Y Lee
    Department of Ophthalmology, University of Washington, Seattle, Washington.