Deep Learning for the Diagnosis of Stage in Retinopathy of Prematurity: Accuracy and Generalizability across Populations and Cameras.

Journal: Ophthalmology. Retina
PMID:

Abstract

PURPOSE: Stage is an important feature to identify in retinal images of infants at risk of retinopathy of prematurity (ROP). The purpose of this study was to implement a convolutional neural network (CNN) for binary detection of stages 1, 2, and 3 in ROP and to evaluate its generalizability across different populations and camera systems.

Authors

  • Jimmy S Chen
    Department of Ophthalmology, Oregon Health and Science University, Portland, Oregon.
  • Aaron S Coyner
    Department of Ophthalmology, Oregon Health and Science University, Portland, Oregon.
  • Susan Ostmo
    Department of Ophthalmology, Oregon Health and Science University, Portland, Oregon.
  • Kemal Sonmez
    Department of Ophthalmology, Oregon Health and Science University, Portland, Oregon.
  • Sanyam Bajimaya
  • Eli Pradhan
  • Nita Valikodath
    Department of Ophthalmology and Visual Sciences, Illinois Eye and Ear Infirmary, University of Illinois at Chicago, Chicago, Illinois.
  • Emily D Cole
    Department of Ophthalmology and Visual Sciences, Illinois Eye and Ear Infirmary, University of Illinois at Chicago, Chicago, Illinois.
  • Tala Al-Khaled
    Department of Ophthalmology and Visual Sciences, Illinois Eye and Ear Infirmary, University of Illinois at Chicago, Chicago, Illinois.
  • R V Paul Chan
    Ophthalmology, Illinois Eye and Ear Infirmary, Chicago, IL, United States.
  • Praveer Singh
    Department of Radiology, MGH/Harvard Medical School, Charlestown, Massachusetts.
  • Jayashree Kalpathy-Cramer
    Department of Radiology, MGH/Harvard Medical School, Charlestown, Massachusetts.
  • Michael F Chiang
    National Eye Institute, National Institutes of Health, Bethesda, Maryland.
  • J Peter Campbell
    Department of Ophthalmology, Oregon Health and Science University, Portland, Oregon.