Table 1.
Configuration of the two OCT instruments.
Figure 1.
Boundaries of intra-retinal layers in OCT macular images.
As seen in this image taken by UHR-OCT in the horizontal meridian, nine boundaries of intra-retinal layers were visualized. Images taken in the vertical meridian by UHR-OCT and in both meridians by the RTVue100 were similar to this.
Figure 2.
The detailed sequence in the boundary segmentation process.
(a) Original image. (b) Image smoothing. (c) Gradient image. (d) The ILM and the boundary between the RPE and choroid layers were first segmented. (e) Limiting detection area and search the minimum-weighted path. (f) Segmented image.
Figure 3.Thickness.
profiles of eight intra-retinal layers determined from the UHR-OCT and RTVue100 images in the horizontal meridian.
Thickness profiles of eight intra-retinal layers along the horizontal meridian were averaged for 20 normal healthy eyes. Error bars represent standard deviation.
Figure 4.
Thickness profiles of eight intra-retinal layers determined from the UHR-OCT and RTVue100 images in the vertical meridian.
Thickness profiles of eight intra-retinal layers along the vertical meridian were averaged for 20 normal healthy eyes. Error bars represent standard deviation.
Figure 5.
Segmentation errors in scans of lower image quality and corresponding corrected segmentation after applying the semi-automated approach.
(A) The algorithm mistakenly identified the OPL/ONL interface. (B) Corrected segmentation corresponding (A) after applying the semi-automated approach. (C) The algorithm mistakenly identified the RNFL/GCL boundary. (D) Corrected segmentation corresponding (C) after applying the semi-automated approach.
Table 2.
Repeatability and reproducibility of thickness measurements for intra-retinal layers measured by UHR-OCT.
Table 3.
Repeatability and reproducibility of thickness measurements for eight intra-retinal layers measured by the RTVue100.
Figure 6.
Bland-Altman plots of thickness measurements determined with the automated segmentation algorithm on UHR-OCT and RTVue100 images.
Only the images along the horizontal meridian were analyzed. The horizontal full lines represent the mean of thickness differences, and the horizontal dashed lines represent the mean differences ±1.96 standard deviation.
Table 4.
Comparison of the thickness measurements for eight intra-retinal layers and total retina between UHR-OCT and RTVue100 devices.