AI technology improves early detection of severe eye inflammation, new study shows

Age-related macular degeneration (AMD) is the leading cause of vision loss in the United States, affecting 11 million people, mostly older adults. The more severe form, neovascular age-related macular degeneration (nAMD), is characterized by abnormal blood vessel growth under the retina. These blood vessels leak fluid and blood, leading to vision loss. In addition to age, smoking, poor diet, and lack of exercise also contribute to the risk of vision loss.

The main treatment for nAMD is anti-VEGF drugs, which involve injecting drugs into the eye to block a protein called vascular endothelial growth factor (VEGF), which is responsible for abnormal blood vessel growth in the retina, but can cause eye inflammation as a serious side effect.

A research team from Emory AI.Health and Cleveland Clinic aimed to predict which patients are likely to develop this inflammatory response. By combining regular optical coherence tomography (OCT) scans with machine learning and precision medicine, they sought to identify patterns in eye scans that may appear before or during inflammation caused by anti-VEGF drugs.

Identifying these patterns early can enable doctors to detect inflammation early and adjust treatment to prevent vision loss.

Spotting problems early: research

Published in HelyonThe study analyzed images from 67 eyes from a retrospective clinical trial of patients with nAMD. The researchers extracted specific texture-based features from the OCT scans, focusing on the vitreous, a clear gel inside the eye. Using a machine learning model developed by Emory AI.Health, they identified patterns indicative of inflammation before it becomes clinically visible.

The machine learning model accurately identified which patients would develop inflammation, achieving 76% accuracy before anti-VEGF treatment and 81% accuracy at the time of injection, suggesting that it may be a valuable tool for early detection.

“Macular degeneration is close to home for me because my father has macular degeneration, and as our population ages, more people will experience nAMD. While anti-VEGF agents can slow the progression of macular degeneration, they also come with risks,” said Anant Madhavshi, PhD, executive director of Emory AI.Health and principal investigator on the study.

“Our study provides valuable data to help clinicians make better treatment decisions, such as reducing dosage or combining it with anti-inflammatory drugs to prevent serious complications.”

“This study validates our AI algorithm in a retrospective clinical trial and highlights the potential of precision medicine in ophthalmology,” said Sudeshna Sir Kar, PhD, first author of the study and an associate scientist at Emory AI.Health. “Next, we hope to incorporate our algorithm into prospective clinical trials to identify patients at risk of developing these adverse events in real time.”