Purchase this article with an account.
Jing Tian, Nathan Bates, Gabor Mark Somfai, Delia DeBuc; Inter-Session Repeatability of Retinal Layer Thickness Measurements in Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci. 2016;57(12):1654.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Reliable retinal layer thickness measurements using optical coherence tomography (OCT) are important to track the subtles changes of the retina in longitudinal studies. We aim to study the inter-session repeatability and identify the pitfalls affecting the reliability of our custom-built segmentation algorithm.
We enrolled 10 eyes of 5 healthy subjects with a mean age of 40±13 years. Each eye was scanned using Spectralis SD-OCT and altogether 3 sessions were taken with a 30s rest in between each. The first and second sessions were scanned independently while the third session was scanned with the first one set as the baseline visit. The first, second and third sessions were named as baseline, unregistered and registered sessions; respectively. Seven retinal layers labeled were segmented using OCTRIMA3D and measured in the ETDRS regions. The inter-session standard deviation (σ), coefficient of repeatability (CR) and coefficient of variation (COV) values were calculated to quantify the repeatability. Paired t-test of COVs was used to compare the repeatability and the level of significance was set at 5%.
In the unregistered session the vessels in the cSLO were rotated and the scanning locations were misaligned. The registered scanning setting allowed the precise alignment of the scanning locations. Inter-session repeatability results obtained in the registered scanning session for all intraretinal layers are shown in Table 1. The values of COV were not significantly different in the unregistered scanning session. The COVs were significantly different (p<0.05) for the OPL, ONL and IS. The OPL-ONL surface was not reliably detected due to the typical isoreflectivity of the Henle fiber layer (HFL). The deviation of the IS and OS measurements was greater than 5% due to the typical thinner average thickness of these layers.
Our results show that the rotations in the unregistered scanning sessions do not cause significant changes in repeatability. The surface between the OPL and ONL has a larger inter-session difference due to the image artifacts created by the HFL isoreflectivity. The measurements of the IS and OS were less reliable due to the typical thinner average thickness of these layers.The combined IS+OS and OPL+ONL segments may yield more reliable measurements.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
Table 1: Repeatability of 7 intraretinal layers in the ETDRS regions of registered scanning sessions.
This PDF is available to Subscribers Only