The National Eye Institute has awarded a $423,500 grant to support groundbreaking research aimed at improving the diagnosis and treatment of a serious eye condition known as uveitis. Siamak Yousefi, PhD, associate professor at the University of Tennessee Health Science Center and fellow of the Association for Research in Vision and Opthalmology (ARVO), will lead this innovative project titled “A Generalizable Deep Learning Platform for Unifying Quantification of Experimental Autoimmune Uveitis.”
Uveitis, particularly posterior uveitis, is responsible for up to 15% of severe vision impairments in the United States. This condition often affects individuals during their childhood or young adulthood and is as impactful as diabetes and macular degeneration in terms of long-term visual impairment. Despite its severity, the causes of uveitis remain largely unknown.
Dr. Yousefi’s research will use a disease model called Experimental Autoimmune Uveitis (EAU), which mirrors essential characteristics of human uveitis. EAU is crucial for studying the mechanisms of uveitis and testing new treatments. However, current methods of assessing EAU are inconsistent, making it difficult to accurately determine the severity of the disease and evaluate potential therapies.
Artificial Intelligence (AI), particularly deep learning, has shown promise in the field of eye health. However, its potential in studying uveitis and EAU specifically is underexplored. Dr. Yousefi’s team aims to develop AI tools that can enhance the accuracy and consistency of diagnosing and grading the severity of uveitis.
The project will focus on creating high-quality datasets that include images from EAU models, captured using fundus photography and optical coherence tomography. The team will then develop AI models to evaluate uveitis severity across five levels. These models will be rigorously tested to ensure they are reliable and applicable to a wide range of cases.
“Our AI-driven approach to experimental uveitis represents a transformative step forward, enabling more precise and uniform quantification of disease severity,” Dr. Yousefi said. “This project will enhance the accuracy, consistency, and speed of uveitis research, paving the way for novel treatments and drug discovery.”
This initiative is a collaborative effort, bringing together experts from various fields, with access to extensive uveitis datasets previously generated at the National Institutes of Health. The outcomes of this research, including datasets and AI tools, will be published in leading scientific journals and made publicly available to advance the vision research community.