Creating realistic anterior segment optical coherence tomography images using generative adversarial networks.

Journal: The British journal of ophthalmology
PMID:

Abstract

AIMS: To develop a generative adversarial network (GAN) capable of generating realistic high-resolution anterior segment optical coherence tomography (AS-OCT) images.

Authors

  • Jad F Assaf
    Faculty of Medicine, American University of Beirut (J.F.A., C.Z., P.B., M.C.), Beirut, Lebanon.
  • Anthony Abou Mrad
    Faculty of Medicine, American University of Beirut, Beirut, Lebanon.
  • Dan Z Reinstein
    London Vision Clinic, London, United Kingdom; Department of Ophthalmology, Columbia University Medical Center, New York, New York, USA; Centre Hospitalier National d'Ophtalmologie, Paris, France; Biomedical Science Research Institute, Ulster University, Coleraine, United Kingdom.
  • Guillermo Amescua
    Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, FL; and.
  • Cyril Zakka
    Faculty of Medicine, American University of Beirut (J.F.A., C.Z., P.B., M.C.), Beirut, Lebanon.
  • Timothy J Archer
    London Vision Clinic, London, United Kingdom.
  • Jeffrey Yammine
    Faculty of Medicine, American University of Beirut, Beirut, Lebanon.
  • Elsa Lamah
    Faculty of Medicine, American University of Beirut, Beirut, Lebanon.
  • Michèle Haykal
    Faculty of Medicine, Saint Joseph University, Beirut, Lebanon.
  • Shady T Awwad
    Department of Ophthalmology, American University of Beirut Medical Center (P.I., S.T.A.), Beirut, Lebanon. Electronic address: sawwad@gmail.com.