Computerized analysis of the eye vasculature in a mass dataset of digital fundus images: the example of age, sex and primary open-angle glaucoma

Publication Lirot.ai Website PVBM toolbox

We are glad to annonce the publication of our latest paper in Ophthalmology Science.

Our study aims to develop and validate an automated end-to-end methodology for analyzing retinal vasculature in large datasets of digital fundus images (DFIs), with a specific focus on assessing the influence of demographic and clinical factors on retinal microvasculature.

This retrospective cohort study analyzed 32,768 digital fundus images from routine eye examinations, employing a comprehensive approach that includes image quality assessment, optic disc detection, region of interest definition, automated segmentation of retinal arterioles and venules, and the engineering of digital biomarkers to represent vasculature characteristics.

Overview of the experiments. The quality of the digital fundus images (DFI) is first automatically computed using FundusQ-Net 30. DFI of sufficient quality are processed with LUNet 29 for arteriole/venule (A/V) and optic disc segmentation. A region of interest is defined around the optic disc to standardize the analysis, and a set of digital vasculature biomarkers is computed using the PVBM toolbox 31. Finally, a statistical analysis of the vasculature characteristics is performed as a function of demographic and clinical variables. Explore our Lirot.ai application to access those features (https://www.aimlab-technion.com/lirot-ai)

Our primary outcomes were changes in retinal vascular geometry, with particular attention to the effects of age, sex, disc size, and primary open-angle glaucoma (POAG). The results revealed significant independent similarities in retinal vascular geometry alterations associated with both advanced age and POAG, suggesting a potential mechanism of accelerated vascular aging in POAG patients.

This novel methodology enables a comprehensive and quantitative analysis of retinal vasculature, providing valuable insights into the state of retinal vascular health and its broader implications for cardiovascular and ocular health. By facilitating reproducible analysis of extensive datasets, our approach offers a critical tool for ongoing and future studies in retinal vasculature.

The software developed through this research is accessible via our Lirot.ai application (https://www.aimlab-technion.com/lirot-ai). The application enables blood vessels segmentation and the engineering of the vasculature biomarkers.