Machine learning flags nAMD drug-related eye inflammation

Researchers from the Emory Empathetic AI for Health Institute (Emory AI.Health) and Cleveland Clinic have developed an AI tool to identify potentially serious eye inflammation caused by the use of anti-vascular endothelial growth factor (VEGF) drugs for neovascular age-related macular degeneration (nAMD).

Also called wet macular degeneration, nAMD is the result of abnormal blood vessel growth under the retina. Leakage from these vessels can lead to irreversible blindness if left untreated.

The research team indicated that the use of anti-VEGF drugs can curb the growth of these blood vessels, but the treatment can also lead to potentially serious cases of intraocular inflammation. To predict which patients might experience this inflammatory response, the researchers sought to develop an AI model capable of flagging intraocular inflammation patterns in eye images.

To develop the tool, the researchers utilized machine learning (ML) to assess routine optical coherence tomography (OCT) scans taken of the vitreous compartment -- the gel-like substance in the eye -- before and during anti-VEGF treatment.

Using OCT scans from 67 eyes of nAMD patients participating in a retrospective clinical trial, the research team tested the model's ability to identify inflammatory patterns before they were clinically visible.

The analysis revealed that the ML tool accurately flagged which patients would develop inflammation 76% of the time prior to anti-VEGF treatment and 81% of the time following treatment.

The research team emphasized that these findings could be valuable for improving nAMD treatment in the future.

"Macular degeneration is personal to me because my father suffers from it. As our population ages, more people will experience nAMD. Anti-VEGF agents can slow down macular degeneration but come with risks," said Anant Madabhushi, Ph.D., executive director of Emory AI.Health and principal investigator of the study, in the news release. "Our study provides valuable data for clinicians to make better treatment decisions, potentially reducing the dosage or combining these agents with anti-inflammatory drugs to prevent severe complications."

Further, the success of the ML model demonstrates the potential of predictive analytics to prevent adverse outcomes.

"This study validates our AI algorithms in a retrospective clinical trial and underscores the potential of precision medicine in ophthalmology," stated Sudeshna Sil Kar, Ph.D., first author of the study and associate scientist at Emory AI.Health. "Next, we hope to embed our algorithms in prospective clinical trials to identify patients likely to develop these adverse events in real-time."

Shania Kennedy has been covering news related to health IT and analytics since 2022.