Improved retinal vascular analysis for accurate detection of pathological changes

Project summary

The retinal vasculature is very sensitive to the effect of several ophthalmic and systemic diseases, which affect a significant portion of the Dutch population, leading to increased healthcare costs and reduced quality of life. Cardiovascular diseases alone account for approximately 29% of deaths in the Netherlands, highlighting the urgent need for effective early diagnostic tools. Given that retinal vascular imaging is non-invasive, fast, and inexpensive, interest in using the retina as a window to cardiovascular health is rapidly increasing.

Optical Coherence Tomography Angiography (OCTA) provides detailed imaging of retinal vasculature; however, analysing this data poses challenges due to frequent acquisition artefacts and the lack of automatic measurement tools. This project seeks to address these limitations by developing robust, open-source methods for biomarker quantification of real-world OCTA data, ensuring they are vendoragnostic and annotation-free.

The project combines Erasmus MC’s expertise in image processing and Artificial Intelligence (AI) for ophthalmic imaging, including OCTA, with Imagine Eyes’ advanced hardware developments. This partnership will provide access to all stages of OCTA image acquisition, from raw 3D data to biomarkers. We will use high-quality 3D data to simulate, detect, and correct artefacts, explore novel, label-free data-driven segmentation techniques, and validate these algorithms in clinical settings. Data from the Faculty of Medicine of the University of Porto will include two cardiovascular disease cohorts, offering a unique retinal OCTA dataset along with comprehensive imaging, clinical, and analytical parameters, allowing for meaningful correlations.

The resulting methods will significantly enhance the usability of OCTA data, enabling earlier diagnosis, improved disease monitoring, and better patient outcomes. By providing widely accessible tools, this project has the potential to lower healthcare costs, support personalized care, and address unmet needs in both ophthalmic and systemic disease diagnostics.