Communications of the ACM
Algorithm Aids in Early Detection of Age-Related Eye Disease
The deep learning algorithm analyzes detailed eye scans to predict if the common eye condition will progress to a more severe stage known as geographic atrophy.
Credit: Duke University School of Medicine
A deep learning algorithm developed by researchers at Duke University can predict whether an individual's age-related macular degeneration (AMD) will progress to geographic atrophy (GA), a more severe stage, within a year.
The DeepGAze algorithm could enable more proactive treatment and better patient outcomes.
In a study of 417 patients, DeepGAze analyzed spectral-domain optical coherence tomography scans of the retina to identify which eyes would progress from intermediate stage AMD to GA within a year, with an accuracy rate of 94%.
The researchers also found the algorithm worked as well on its own as when expert graders added image annotations to indicate certain conditions.
Duke's Dr. Eleonora Lad said, "This predictive tool could transform how we screen for the disease, how often we see patients, and even when to start treatments."
From Duke University School of Medicine
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found