Abstract
Purpose:
Alterations in the choroid play an important role in the pathogenesis of age-related macular degeneration (AMD). Manual and semi-automated methods of segmentation have shown good discrimination of choroidal detail on optical coherence tomography (OCT) but fully automated protocols to detect outer retinal structures in AMD are lacking. This was a pilot study, in which we applied a newly developed method of automated OCT segmentation analysis to study the correlations of choroid, choriocapillaris, retina, and RPE thickness in eyes with AMD.
Methods:
38 subjects with AMD had 61 line scan protocol with enhanced depth imaging performed on the Heidelberg Spectralis SD-OCT device. Automated segmentation using a graph-based method was used to determine thickness of photoreceptor outer segment length, RPE, choriocapillaris, and choroid. Scans of eyes with CNV were excluded from analysis. These measurements were plotted and a linear regression analysis was used to assess for statistical significance.
Results:
The mean thickness (μ) + standard deviation (SD) of the choriocapillaris, RPE, OSL, and choroid was 8.73+0.65; 24.2+1.00; 39.02+6.42; 158.21+52.21 respectively. The strongest correlation (r2=0.200) was found between the thickness of the choroid and the OSL. There was also weak correlation between the RPE and the OSL (r2=0.121). All other relationships had an r2 < 0.100.
Conclusions:
Automated segmentation of SD-OCT images can be successfully performed to determine thickness of outer retinal structures of eyes with AMD. There is a plausible correlation between photoreceptor outer segment length and choroid that needs confirmation in a larger data set.