Predicting Treat-and-Extend Outcomes and Treatment Intervals in Neovascular Age-Related Macular Degeneration from Retinal Optical Coherence Tomography Using Artificial Intelligence

Published

2022

Audience

Therapeutic Area

Content Type

AI Generated Summary*

Early optical coherence tomography scans were analyzed retrospectively to forecast visual response and retreatment needs within the treat-and-extend arm of the TREND ranibizumab trial in treatment-naive neovascular age-related macular degeneration. Of 228 evaluable eyes, 55% achieved and sustained intervals of 8 weeks or more, while 45% remained at 4 or 6 weeks. Automated segmentation quantified intraretinal and subretinal fluid, pigment epithelial detachment, hyperreflective foci and photoreceptor layer thickness at baseline and week 4, feeding a cross-validated random forest. Using clinical plus imaging features, the model predicted extendable status with an AUC of 0.71 and responder status, defined from BCVA trajectories, with 0.87. Among imaging features, residual subretinal fluid volume best predicted intervals and intraretinal fluid best predicted vision.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.

At a glance

Authors
Hrvoje Bogunović; Virginia Mares; Gregor S. Reiter; Ursula Schmidt-Erfurth
Publication
Frontiers in Medicine
Keywords
Artificial Intelligence, Anti-VEGF, Image Analysis, Machine Learning, Neovascular Age Related Macular Degeneration, Optical Coherence Tomography, Retina
Year
2022
View primary source

This is the evidence

See where we take it next

Join the Network