Physics-informed deep generative learning for quantitative assessment of the retina.

Journal: Nature communications
Published Date: Aug 10, 2024

Abstract

Disruption of retinal vasculature is linked to various diseases, including diabetic retinopathy and macular degeneration, leading to vision loss. We present here a novel algorithmic approach that generates highly realistic digital models of human retinal blood vessels, based on established biophysical principles, including fully-connected arterial and venous trees with a single inlet and outlet. This approach, using physics-informed generative adversarial networks (PI-GAN), enables the segmentation and reconstruction of blood vessel networks with no human input and which out-performs human labelling. Segmentation of DRIVE and STARE retina photograph datasets provided near state-of-the-art vessel segmentation, with training on only a small (n = 100) simulated dataset. Our findings highlight the potential of PI-GAN for accurate retinal vasculature characterization, with implications for improving early disease detection, monitoring disease progression, and improving patient care.

Authors

  • Emmeline E Brown
    Centre for Computational Medicine, University College London, London, UK.
  • Andrew A Guy
    Centre for Computational Medicine, University College London, London, UK.
  • Natalie A Holroyd
    Centre for Computational Medicine, University College London, London, UK.
  • Paul W Sweeney
    Cancer Research UK Cambridge Institute, University of Cambridge, Robinson Way, Cambridge, CB2 0RE, UK.
  • Lucie Gourmet
    Centre for Computational Medicine, University College London, London, UK.
  • Hannah Coleman
    Centre for Computational Medicine, University College London, London, UK.
  • Claire Walsh
    Centre for Computational Medicine, University College London, London, UK.
  • Athina E Markaki
    Department of Engineering, University of Cambridge, Cambridge, UK.
  • Rebecca Shipley
    Centre for Computational Medicine, University College London, London, UK.
  • Ranjan Rajendram
    Moorfields Eye Hospital, London, UK.
  • Simon Walker-Samuel
    Centre for Computational Medicine, University College London, London, UK. simon.walkersamuel@ucl.ac.uk.

Keywords

External Resources

Popular Topics
Recent Journals