top of page

RETFOUND : Preventing Blindness Worldwide with AI

According to the World Health Organization (WHO), at least 2.2 billion people have a near or distance vision impairment. In many cases, these are preventable, yet the tools for early detection are not available to all.


Blindness and vision impairment present not just a health challenge but a significant impediment to economic and social participation.


RETFOUND stands as the world's first artificial intelligence foundation model tailored for eye care, a groundbreaking development in the detection of eye diseases and the prediction of systemic health issues. Developed by researchers at University College London (UCL), this AI system can analyze eye scans with unprecedented accuracy, identifying signs of diseases that even seasoned professionals might miss. RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels.


Beyond mere detection, RETFOUND’s ability to foresee the potential systemic health problems linked to eye conditions positions it as a tool for holistic health. By integrating this AI with routine eye exams, RETFOUND transforms a simple check-up into a comprehensive health evaluation.


This pioneering technology is set to supercharge global efforts in preventing blindness, particularly in underserved regions where access to specialists is limited. By democratizing diagnosis, RETFOUND empowers local healthcare providers, bridges gaps in care, and paves the way for early intervention.


The inception of RETFOUND is a pivotal moment where technology doesn't just support healthcare; it redefines it. As this AI model learns and grows, it holds the promise of a future where blindness prevention is a universal reality, and every set of eyes receives the care it deserves.


We invite you to join our Unicorns for Good world-class community of impact leaders to co-create a more mindful, inclusive, and sustainable world, by signing our Unicorns for Good Pledge.


 
 
 

Comments


bottom of page