Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Apr 7;5(5):100787.
doi: 10.1016/j.xops.2025.100787. eCollection 2025 Sep-Oct.

Longitudinal Comparison of Geographic Atrophy Enlargement Using Manual, Semiautomated, and Deep Learning Approaches

Jacob Bogost  1 , Rachel E Linderman  1   2 , Robert Slater  1   2 , Thomas F Saunders  1   2 , Caleb Pacheco  2 , Jeong Pak  1   2 , Rick Voland  2 , Barbara Blodi  2 , Amitha Domalpally  1   2
Affiliations

Longitudinal Comparison of Geographic Atrophy Enlargement Using Manual, Semiautomated, and Deep Learning Approaches

Jacob Bogost et al. Ophthalmol Sci. .

Abstract

Objective: To compare a fully automated artificial intelligence (AI) model, a semiautomated method, and manual planimetry in the longitudinal assessment of geographic atrophy (GA) using fundus autofluorescence images.

Design: A retrospective analysis of 3 GA assessment methods: AI, Heidelberg Eye Explorer semiautomated software (RegionFinder), and manual planimetry.

Subjects and controls: One hundred eight patients (185 eyes) with GA from a phase IIb clinical trial by GlaxoSmithKline, which evaluated an experimental drug that did not reduce GA enlargement compared with the placebo.

Methods: Fundus autofluorescence images of 185 eyes were annotated using manual planimetry, semiautomated RegionFinder, and a fully automated AI model trained and validated on manual planimetry annotations at screening, year 1, and year 2. Artificial intelligence masks were compared with human-guided methods, and regression errors were assessed by stacking masks from consecutive visits. Agreement between methods was assessed using Bland-Altman plots, Dice similarity coefficient (DSC), and comparisons of GA growth rates. Artificial intelligence performance was evaluated based on its need for human edits and frequency of regression errors.

Main outcome measures: Agreement between methods was evaluated using Bland-Altman plots, DSC, and intraclass correlation coefficients (ICCs). The mean GA growth rate (mm2/year) and square root transformation of GA size were compared across methods. Artificial intelligence performance was assessed by the percentage of acceptable masks and the frequency of longitudinal regression errors.

Results: At screening, the mean GA area was 7.22 mm2 with RegionFinder, 8.37 mm2 with AI, and 8.66 mm2 with manual planimetry. RegionFinder measured smaller GA areas than both AI and manual, with a mean difference of -1.45 mm2 (95% confidence interval [CI]: -1.56, -1.35) versus AI (ICC = 0.945) and -1.87 mm2 (95% CI: -1.99, -1.75) versus manual (ICC = 0.920). Growth rates were comparable between RegionFinder (1.54 mm2/year), AI (1.68 mm2/year), and manual (1.80 mm2/year) (P = 0.25). Artificial intelligence masks were deemed acceptable in 84.8% of visits, and 81.4% of cases showed no regression over time.

Conclusions: Artificial intelligence accurately measures GA in approximately 85% of cases, requiring human intervention in only 15%, indicating potential to streamline GA measurement in clinical trials while maintaining human oversight.

Financial disclosures: The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Keywords: Artificial intelligence; Fundus autofluorescence; Geographic atrophy; RegionFinder; Retinal imaging.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Measurement of GA area over 2 years with 3 methods: RegionFinder, AI, and manual perimetry. The GA is heterogeneous, as seen by lighter areas within the hypo-autofluorescent lesion. Both RegionFinder and AI methods exclude these lighter areas, and the manual method includes them resulting in a larger area and slower enlargement. AI = artificial intelligence; GA = geographic atrophy.
Figure 2
Figure 2
Analysis of longitudinal regression. The top row shows the raw image and corresponding AI mask at screening, while the bottom row displays the raw image and AI mask at year 1. The rightmost composite image represents the superimposed masks from both visits, where yellow indicates consistent GA area, red signifies regression, and green represents growth. Examples are provided for each grading category: no regression (A), minor regression (B), and major regression (C). In panel C, the follow-up image misses GA foci due to suboptimal image quality. Instances where GA foci were missed at year 1 were classified as major regressions. AI = artificial intelligence; GA = geographic atrophy.
Figure 3
Figure 3
Bland−Altman plots for comparison of the GA area between RegionFinder and AI (A) and RegionFinder and manual planimetry (B). AI = artificial intelligence; GA = geographic atrophy.
Figure 4
Figure 4
Comparison of change in area of GA over 2 years as measured by RegionFinder, AI, and manual planimetry. Mean growth rate was 1.54 mm2/year, 1.68 mm2/year, and 1.80 mm2/year, respectively. AI = artificial intelligence; GA = geographic atrophy.
See this image and copyright information in PMC

References

    1. Fleckenstein M., Schmitz-Valckenberg S., Chakravarthy U. Age-related macular degeneration: a review. JAMA. 2024;331:147–157. - PubMed
    1. Boyer D.S., Schmidt-Erfurth U., van Lookeren Campagne M., et al. The pathophysiology of geographic atrophy secondary to age-related macular degeneration and the complement pathway as a therapeutic target. Retina. 2017;37:819–835. - PMC - PubMed
    1. Wang J., Ying G.S. Growth rate of geographic atrophy secondary to age-related macular degeneration: a meta-analysis of natural history studies and implications for designing future trials. Ophthalmic Res. 2021;64:205–215. - PubMed
    1. Holz F.G., Bellman C., Staudt S., et al. Fundus autofluorescence and development of geographic atrophy in age-related macular degeneration. Invest Ophthalmol Vis Sci. 2001;42:1051–1056. - PubMed
    1. Heier J.S., Lad E.M., Holz F.G., et al. Pegcetacoplan for the treatment of geographic atrophy secondary to age-related macular degeneration (OAKS and DERBY): two multicentre, randomised, double-masked, sham-controlled, phase 3 trials. Lancet. 2023;402:1434–1448. - PubMed

LinkOut - more resources