AI Taught Physics Brings Eye Disease Into Sharper Focus

Less noise in sight.

University of Waterloo engineers have developed physics-trained AI that improves medical imaging of the eye, allowing for earlier and more accurate detection of diseases like glaucoma and macular degeneration.

While traditional deep-learning systems rely solely on data, this new model integrates the physics of light and tissue into its learning process. This allows for the reconstruction of clear, high-resolution images from noisy scans, ultimately reducing imaging costs for medical clinics.

“By merging the power of AI with the knowledge of physics, our model can methodically reduce such errors and produce more trustworthy results.”

— Dr. Alexander Wong

Read more →