New Study Confirms AI’s Reliability in Home OCT Analysis for nAMD

New Study Demonstrates AI’s Reliability in Home OCT Analysis for nAMD 

Can AI finally match human experts in analyzing home OCT scans for nAMD? A new study says yes.

Notal Vision (Manassas, Virginia, USA) has announced the publication of a pivotal study in Ophthalmology Science that evaluated the performance of an AI algorithm in analyzing key biomarkers for nAMD using a home-based optical coherence tomography (OCT) system.1

The recently published study contains the detailed data that led to the first-ever FDA clearance for an AI-driven algorithm applied to OCT images—a major leap forward in remote patient monitoring.

“This is one of the most significant studies related to artificial intelligence and retinal disease management as it led to clearance of the first-ever AI for OCT,” said Eric Schneider, MD of Tennessee Retina and senior author of the study. “It should give retina specialists a great deal of confidence in adopting this patient monitoring program.”

The findings of the study will be presented at the Bascom Palmer Eye Institute’s (Miami, Florida, USA) virtual Angiogenesis, Exudation, and Degeneration 2025 meeting on February 8 by renowned retinal specialist Prof. Anat Loewenstein (Israel), professor and head of the Department of Ophthalmology at the Tel Aviv Medical Center.

For nAMD patients, frequent monitoring is crucial to track disease activity and treatment response. However, with the high volume of OCT images generated by at-home scans, manual review simply isn’t feasible. That’s where AI comes in—offering a way to process and analyze images in real-time, flagging important changes for physicians. 

But can AI be as reliable as expert human graders? This new study says yes.

Putting AI to the test: The study breakdown

The cross-sectional study involved 336 nAMD patients from six retina clinics across the U.S. Each patient captured images using two home OCT devices, without technician assistance, in a controlled clinic setting. They also underwent in-office OCT imaging for comparison.

The AI algorithm analyzed hypo-reflective spaces (HRS)—fluid spaces in the retina that indicate disease activity—and calculated total retinal hypo-reflective space (TRO) volumes. The study was aimed at finding answers to these two questions: 

  • Repeatability: Could the AI algorithm consistently measure TRO volumes across multiple home OCT scans?
  • Accuracy: How well did the AI segmentation of HRS match up with expert graders?

AI matches expert graders in home OCT analysis

The findings were promising. The AI algorithm not only demonstrated high repeatability in its measurements but also matched expert graders in identifying and segmenting fluid spaces. In fact, the AI’s performance was found as reliable as the agreement between expert graders themselves.

“The positive results enable SCANLY Home OCT, the first market-authorized home imaging device in eyecare, to provide physicians with actionable insights in disease activity and treatment response between patients’ office visits,” said Kester Nahen, PhD, CEO of Notal Vision.

In May 2024, SCANLY Home OCT received FDA De Novo authorization, making it the first self-operated imaging device for patients with wet age-related macular degeneration (AMD). Designated as a breakthrough device, it aims to improve personalized disease monitoring and management.

The device captures spectral-domain optical coherence tomography (OCT) images within a 10×10-degree area centered on the fixation point. Powered by the Notal OCT Analyzer (NOA), an AI-driven system, it segments and quantifies hypo-reflective spaces (HRS), key biomarkers used in AMD management.

Reference 

  1. Schneider EW, Heier JS, Holekamp NM, et al.  Pivotal Trial Toward Effectiveness of Self-administered OCT in Neovascular Age-related Macular Degeneration. Report 2-Artificial Intelligence Analytics. Ophthalmol Sci. 2024;5(2):100662.
Subscribe
Notify of
guest
0 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments