Abstract
Purpose:
To evaluate the repeatability and reproducibility of thickness profile measurements of eight intra-retinal layers with a newly developed automatic segmentation using ultra-high resolution optical coherence tomography (UHR-OCT). The agreement of these automatic thickness profiles in normal eyes was also investigated between UHR-OCT and RTVue OCT.
Methods:
Twenty normal subjects (aged from 24 to 32 years, mean 25.1±2.0) and five glaucoma patients (aged from 35 to 75 years, mean 54.5±13.3) were enrolled. A custom built UHR-OCT (~ 3μm) and a commercial RTVue OCT (~ 5μm) were used to image retina. An automatic segmentation algorithm was developed to obtain the thickness profiles of eight intra-retinal layers including RNFL, GCL+IPL, INL, OPL, ONL, IS, OS and RPE. Right eye in each normal subject was imaged both in the horizontal and vertical meridians. The retinal images of each eye were taken three times by two examiners by UHR-OCT and taken one time by RTVue OCT. The thickness profiles of eight intra-retinal layers for glaucoma patients were compared with those for normal subjects.
Results:
The coefficients of repeatability and reproducibility were less than 5.04% for the total retinal thickness and all eight intra-retinal layers. The intraclass correlation coefficient (ICC) for IS was 0.34, while the ICC for the other seven intra-retinal layers and total retinal thickness were greater than 0.94. The ICC for agreement of the thickness profiles between UHR-OCT and RTVue OCT were greater than 0.80 except the layers of IS and OS. The total retinal thickness (273.6μm), RNFL (17.9μm) and GCC+IPL (47.9μm) of glaucoma were found to be thinner than that in normal eye (293.0μm, 27.71μm, 64.8μm, respectively, P < 0.05). No significant differences were found in other six intra-retinal layers between these two groups (P > 0.05).
Conclusions:
The good repeatability, reproducibility and agreement with the commercial OCT indicated that UHR-OCT with the automatic segmentation algorithm is reliable to measure the intra-retinal thickness profiles and may be a promising tool to diagnosis the retinal diseases, such as glaucoma.
Keywords: 550 imaging/image analysis: clinical •
688 retina