Purchase this article with an account.
Andrew M. Huang, Robert F Cooper, Grace Vergilio, Jean Bennett, Albert M. Maguire, Tomas S Aleman, Jessica Ijams Wolfing Morgan; Cone Outer Segment Reflectance Entropy in Choroideremia. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1030. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Adaptive optics scanning light ophthalmoscopy (AOSLO) has previously shown near normal or reduced cone density in patients with choroideremia (CHM). Despite this, abnormal cone outer segment reflectance patterns are readily observed in confocal AOSLO images from these same patients. Here, we use reflectance entropy as a metric to quantify cone reflectance and compare cone reflectance entropy between CHM patients and normal-sighted controls.
The photoreceptor mosaics of 14 CHM and 12 control subjects were imaged using a custom, multi-modal AOSLO at retinal locations surrounding the fovea and extending to either 8° or the atrophic CHM border, whichever was closer, along all four retinal meridians. Images were montaged and regions of interest (ROIs) showing the cone mosaic were selected along the temporal arm at 0.5, 1, 2, 4 and 8°. Cones were identified manually/semi-automatically from split detection and confocal images in CHM patients/controls, respectively, using custom software. Bound cone density and nearest neighbor distance (NND) were calculated from the cone locations for each ROI. Cone reflectance entropy was calculated using a gray-level co-occurrence matrix over the area encompassed by each cone, where each cone’s area was defined as a circle centered on the cone’s location with a radius that was 40% of the ROI’s mean NND. An unequal variance T-test was used to evaluate differences in cone density and reflectance entropy between CHM and controls.
A total of 12,795 CHM and 20,427 normal cones were analyzed. Consistent with prior reports, cone density was reduced in CHM compared to normal (p=0.03). Composite cone reflectance entropy was lower in CHM than in controls (2.2 vs 2.7, respectively; p<0.001). By location, CHM entropies (mean±stdev) were 2.1±0.3, 2.2±0.4, 2.3±0.4, 2.2±0.4, and 2.4±0.3 at 0.5, 1, 2, 4, and 8°, respectively. Control cone reflectance entropies were: 2.4±0.1, 2.7±0.2, 2.8±0.2, 2.8±0.2, and 2.9±0.1, at 0.5, 1, 2, 4, and 8°, respectively. Differences in reflectance entropy between CHM and control were statistically significant (p<0.01) at all locations except 8°(p=0.09).
Cone outer segment reflectance entropy is decreased in patients with CHM. Cone reflectance entropy may prove useful for quantifying changes in the reflectance profiles of cones that remain structurally present in disease.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only