American Academy of Ophthalmology, Oct. 18-20

The annual meeting of the American Academy of Ophthalmology was held from Oct. 18 to 20 in Orlando, Florida, and attracted participants from around the world, including ophthalmologists, optometrists, opticians, and other eye health care professionals. The conference featured presentations focusing on the latest advances in comprehensive eye care, including medical, surgical, and optical care.

In one presentation, Arslon Humayun, an M.D. candidate at the Georgetown University School of Medicine in Washington, D.C., and colleagues discussed the impacts of extended spaceflight on vision and how to address resulting challenges.

The authors highlighted that spaceflight poses significant challenges to the eyes, from microgravity and radiation exposure to chemicals and ocular trauma. Most notable among these challenges is spaceflight-associated neuro-ocular syndrome, or SANS, a cluster of changes that are believed to occur due to microgravity exposure and cephalic fluid shifts. These changes include choroidal folding, globe flattening, optic disc edema, high intracranial pressure without symptoms of intracranial hypertension, and changes to the optic nerve and brain elasticity.

“We are entering a new era of spaceflight, and we need to make sure we are prepared. Ocular challenges in space are well documented from the earliest days of spaceflight, but there are still many unknowns about the effects of spaceflight on vision,” Humayun said. “Further research and additional tools are needed to address these challenges (for example, improved spacesuits to reduce light-flash phenomena resulting from exposure to galactic cosmic rays), particularly as humanity seeks to go deeper into space and on longer spaceflights.”

Press Release

In another study, Anthony Khawaja, Ph.D., of the UCL Institute of Ophthalmology and Moorfields Eye Hospital in London, and colleagues found that artificial intelligence (AI) assessment of the optic nerve vertical cup-disc ratio (VCDR) from retinal photos represents a major improvement over human assessment of photos for glaucoma screening, and this improved performance may help enable cost-effective screening in the future.

Using data from a large, population-based cohort study (the EPIC-Norfolk Eye Study), the authors compiled more than 6,000 fundus images to compare the accuracy of their AI assessment tool (machine learning algorithm) and a trained human grader to estimate the VCDR.

The researchers found that their AI algorithm outperformed humans at grading the VCDR from retinal photos, which in turn provided a more accurate test for glaucoma diagnosis and screening. Specifically, the AI algorithm effectively identified patients with glaucoma 88 to 90 percent of the time, while human graders effectively identified patients with glaucoma 79 to 81 percent of the time.