Abstract
Purpose :
Multiple en face image averaging can be used to improve image quality for optical coherence tomography angiography (OCTA), but little is known regarding its effect on quantification in the radial peripapillary capillary (RPC) layer. We performed a cross-sectional clinical study to test the hypothesis that triplicate averaging of the RPC layer would improve visualization and increase diagnostic accuracy of OCTA in glaucoma.
Methods :
We analyzed 6x6mm triplicate images of the optic nerve head in primary open angle glaucoma (POAG) eyes and glaucoma suspect (GS) eyes using spectral domain-OCTA at the University of Southern California Roski Eye Institute. Images with poor quality or signal strength less than 7 were excluded. Images from the RPC layer were registered using the first image as a reference, and the three frames were then averaged using a custom MATLAB program. Parameters of global entropy (GE), global standard deviation (GSD), local texture correlation (LC), local homogeneity (LH), signal-to-noise ratio (SNR), and intercapillary distance (ICD) were used to measure the change in visualization with averaging. Parameters of vessel density (VD), vessel skeleton density (VSD), vessel complexity index (VCI) and flux were calculated globally and in a 2.8mm annulus, excluding the optic disc, and used to evaluate the change in quantification. Finally, the difference in diagnostic accuracy was assessed using the area under the receiver operating curve (AUROC) for VD in the annulus.
Results :
39 POAG eyes and 76 GS eyes were included. 3-frame averaging resulted in decreased GE and GSD (Ps<0.0001), and increased LC, LH, SNR, and ICD (Ps<0.0001). Averaged images also had decreased VD, VSD, VCI, and flux (Ps<0.0001). A trend towards increased AUROC was observed with VD from the averaged image as compared to the single image (p=0.0577).
Conclusions :
3-frame image averaging significantly improves visualization of the RPC layer and results in higher diagnostic accuracy of OCTA in glaucoma, which suggests that it is advantageous for clinical application. Further studies are needed to assess whether the improvement in diagnostic accuracy is significant and reproducible.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.