Measuring Geographic Atrophy Area Using Column-Based Machine Learning Software on Spectral-Domain Optical Coherence...

Area of Science:

• Ophthalmology
• Medical Imaging
• Retinal Diseases

Background:

• Dry age-related macular degeneration (AMD) is characterized by geographic atrophy (GA).
• Accurate measurement of GA is crucial for monitoring disease progression and treatment efficacy.
• Current imaging modalities like fundus autofluorescence (FAF) and optical coherence tomography (OCT) are used for GA assessment.

Purpose of the Study:

• To compare semi-automatic GA area measurements between FAF and OCT using the cRORA criteria.
• To evaluate differences in GA characteristics (area, focality, perimeter, etc.) derived from FAF versus OCT.

Main Methods:

• Semi-automatic annotation of GA findings on FAF and OCT scans from 36 eyes with dry AMD.
• Comparison of quantitative GA parameters including area, perimeter, and diameters between the two modalities.
• Multivariate regression analysis to identify factors correlating with measurement discrepancies.

Main Results:

• GA area measured on OCT (4.74 ± 3.80 mm²) was significantly smaller than on FAF (13.47 ± 8.64 mm²).
• A significant correlation was found between the difference in GA area measurements and baseline lesion perimeter, maximal diameter on OCT, and total lesion area on FAF.
• Multivariate analysis indicated that lesion characteristics significantly influenced the discrepancy between OCT and FAF measurements.

Conclusions:

• Significant differences exist in GA area measurements obtained via FAF and OCT.
• The choice of imaging modality impacts GA area quantification.
• Further research is needed to establish the clinical significance of these measurement discrepancies.