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

• Published in: JCI insight Vol. 2; no. 24
• Main Authors: Xiao, Sa, Bucher, Felicitas, Wu, Yue, Rokem, Ariel, Lee, Cecilia S., Marra, Kyle V., Fallon, Regis, Diaz-Aguilar, Sophia, Aguilar, Edith, Friedlander, Martin, Lee, Aaron Y.
• Format: Journal Article
• Language: English
• Published: United States American Society for Clinical Investigation 21.12.2017
• Subjects: Algorithms; Animals; Deep Learning; Disease Models, Animal; Female; Image Processing, Computer-Assisted - methods; Male; Mice, Inbred C57BL; Mice, Transgenic; Microscopy, Confocal - methods; Observer Variation; Oxygen; Retinal Neovascularization - diagnosis; Retinal Neovascularization - etiology; Retinal Neovascularization - pathology; Technical Advance; Angiogenesis; Retinopathy; Ophthalmology
• ISSN: 2379-3708; 2379-3708
• DOI: 10.1172/jci.insight.97585

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.