Automatic Quantification Software for Geographic Atrophy Associated with Age-Related Macular Degeneration: A...

Area of Science:

• Ophthalmology
• Medical Imaging
• Computational Biology

Background:

• Age-related macular degeneration (AMD) is a leading cause of vision loss.
• Geographic atrophy (GA) is a key feature of advanced AMD.
• Accurate quantification of GA is crucial for monitoring disease progression and treatment efficacy.

Purpose of the Study:

• To evaluate the accuracy and repeatability of novel software for automated GA area quantification.
• To compare automated measurements with manual assessments using swept-source optical coherence tomography (SS-OCT).

Main Methods:

• Forty-six eyes from 33 AMD patients with GA were scanned using SS-OCT.
• Three masked observers manually measured GA areas from fundus autofluorescence (FAF) images.
• Automated GA area measurements from the software were compared to manual measurements using Lin's concordance correlation coefficient (CCC).

Main Results:

• The overall CCC between manual and automatic GA area measurements was 0.95.
• High agreement was observed for single lesions (CCC=0.99) and regular lesions (CCC=0.97).
• The software demonstrated excellent repeatability with an intraclass correlation coefficient (ICC) of 0.98.

Conclusions:

• The new software provides substantial accuracy and high repeatability for quantifying GA areas in AMD patients.
• The automated tool is particularly precise for measuring single GA lesions.
• This technology has the potential to improve the monitoring of GA progression in clinical practice.