Go to The Journal of Clinical Investigation
  • About
  • Editors
  • Consulting Editors
  • For authors
  • Publication ethics
  • Transfers
  • Advertising
  • Job board
  • Contact
  • Current issue
  • Past issues
  • By specialty
    • COVID-19
    • Cardiology
    • Immunology
    • Metabolism
    • Nephrology
    • Oncology
    • Pulmonology
    • All ...
  • Videos
  • Collections
    • Resource and Technical Advances
    • Clinical Medicine
    • Reviews
    • Editorials
    • Perspectives
    • Top read articles
  • JCI This Month
    • Current issue
    • Past issues

  • Current issue
  • Past issues
  • Specialties
  • In-Press Preview
  • Concise Communication
  • Editorials
  • Viewpoint
  • Top read articles
  • About
  • Editors
  • Consulting Editors
  • For authors
  • Publication ethics
  • Transfers
  • Advertising
  • Job board
  • Contact
Fully automated, deep learning segmentation of oxygen-induced retinopathy images
Sa Xiao, … , Martin Friedlander, Aaron Y. Lee
Sa Xiao, … , Martin Friedlander, Aaron Y. Lee
Published December 21, 2017
Citation Information: JCI Insight. 2017;2(24):e97585. https://doi.org/10.1172/jci.insight.97585.
View: Text | PDF
Resource and Technical Advance Angiogenesis Ophthalmology

Fully automated, deep learning segmentation of oxygen-induced retinopathy images

  • Text
  • PDF
Abstract

Oxygen-induced retinopathy (OIR) is a widely used model to study ischemia-driven neovascularization (NV) in the retina and to serve in proof-of-concept studies in evaluating antiangiogenic drugs for ocular, as well as nonocular, diseases. The primary parameters that are analyzed in this mouse model include the percentage of retina with vaso-obliteration (VO) and NV areas. However, quantification of these two key variables comes with a great challenge due to the requirement of human experts to read the images. Human readers are costly, time-consuming, and subject to bias. Using recent advances in machine learning and computer vision, we trained deep learning neural networks using over a thousand segmentations to fully automate segmentation in OIR images. While determining the percentage area of VO, our algorithm achieved a similar range of correlation coefficients to that of expert inter-human correlation coefficients. In addition, our algorithm achieved a higher range of correlation coefficients compared with inter-expert correlation coefficients for quantification of the percentage area of neovascular tufts. In summary, we have created an open-source, fully automated pipeline for the quantification of key values of OIR images using deep learning neural networks.

Authors

Sa Xiao, Felicitas Bucher, Yue Wu, Ariel Rokem, Cecilia S. Lee, Kyle V. Marra, Regis Fallon, Sophia Diaz-Aguilar, Edith Aguilar, Martin Friedlander, Aaron Y. Lee

×

Full Text PDF | Download (2.34 MB)


Copyright © 2022 American Society for Clinical Investigation
ISSN 2379-3708

Sign up for email alerts