New Eye-Scan Breakthrough Could Make Glaucoma Detection More Reliable

Table of Links

Abstract and 1. Introduction

2. Methods

   2.1 Participants

   2.2 Data Acquisition

   2.3 NFL Reflectance Analysis

3. Results

   3.1 Neutral Density Filter Experiment

   3.2 FSOCT Clinic Data Analysis

4. Discussion

5. Conclusions and References

Abstract

Purpose: Reliability for Nerve Fiber Layer Reflectance Using Spectral Domain Optical Coherence Tomography (OCT)

Methods: The study utilized OCT to scan participants with a cubic 6x6 mm disc scan. NFL reflectance were normalized by the average of bands below NFL and summarized. We selected several reference bands, including the pigment epithelium complex (PPEC), the band between NFL and Bruch's membrane (Post-NFL), and the top 50% of pixels with higher values were selected from the Post-NFL band by Post-NFL-Bright. Especially, we also included NFL attenuation coefficient (AC), which was equivalent to NFL reflectance normalized by all pixels below NFL. An experiment was designed to test the NFL reflectance against different levels of attenuation using neutral density filter (NDF). We also evaluated the within-visit and between-visit repeatability using a clinical dataset with normal and glaucoma eyes.

Results: The experiment enrolled 20 healthy participants. The clinical dataset selected 22 normal and 55 glaucoma eyes with at least two visits form functional and structural OCT (FSOCT) study. The experiment showed that NFL reflectance normalized PPEC Max and Post-NFL-Bright had lowest dependence, slope=- 0.77 and -1.34 dB/optical density on NDF levels, respectively. The clinical data showed that the NFL reflectance metrics normalized by Post-NFL-Bright or Post-NFL-Mean metrics had a trend of better repeatability and reproducibility than others, but the trend was not significant. All metrics demonstrated similar diagnostic accuracy (0.82-0.87), but Post-NFL-Bright provide the best result.

Conclusions: The NFL reflectance normalized by the maximum in PPEC had less dependence of the global attenuation followed by Post-NFL-Bright, PPEC/Mean, Post-NFL-Mean and NFL/AC. But NFL reflectance normalized by Post-NFL-Bright had better result in two datasets.

1 Introduction

Retinal Nerve Fiber Layer (RNFL) thickness, phase retardation, and reflectance are important biomarkers for glaucoma diagnostics [1]. Especially NFL thickness by Optical Coherence Tomography (OCT) has been a well-established method to monitor glaucoma progression and confirmation. However, for mass screening, NFL thickness is not suitable to be used alone [2]. It could be noted, Retinal ganglion cells (RGC) and their axons are significantly damaged by glaucoma which leads to visual field abnormalities and vision loss [1,3]. A study shows that RNFL reflectance is more sensitive to glaucoma damage than a change in RNFL thickness [3]. Further, a decrease in RNFL reflectance occurs before thinning of the RNFL [6]. Hence, an investigation into the RNFL reflectance analysis technique is imperative.

NFL reflectance was derived from OCT image is sensitive to different attenuations due to the optical scattering properties, e.g., media opacity, poor focusing, and ocular media opacities. These effects are often compensated by reference layers [1,4,5]. In [1] and [4], the RNFL reflectivity is normalized by the Retinal Pigment Epithelium (RPE). In [1] and [4], the RNFL reflectivity is normalized by the Retinal Pigment Epithelium (RPE). However, Gardiner SK at el. [5] showed that NFL reflectance corrected by the Post-NFL band had better repeatability than the RPC band. It could be noted, upon observation, we noticed that the Post-NFL band has a darker layer, which could be a disadvantage. However, by selecting the top 50% of pixels, we can opt out of those regions and potentially gain better advantages. Therefore, we selected the top 50% of pixels from the Post-NFL band and named it Post-NFL-Bright. Moreover, we observed that more layers below the NFL provide better repeatability; hence we considered the summation of attenuation coefficients (AC). The attenuation coefficient is an optical property that explains the attenuation of light that occurs due to the scattering and absorption properties of tissue. Therefore, the determination of the attenuation coefficient provides valuable information on glaucoma. Study [3] shows that the average RNFL attenuation coefficients are fully separable for normal and glaucoma. Several papers [4, 6,7,8] have been published for the quantitative analysis of attenuation coefficients. In this depth-resolved [7] based technique was used to determine RNFL attenuation coefficient maps based on a method that uses the retinal pigment epithelium as a reference layer. We have extended the depth-resolved based method by summation of the attenuation coefficient (AC) in NFL layers (which we called NFL/AC) because we want to compare and evaluate the NFL/AC together with other references (NFL/PPEC, NFL/Post-NFL-Mean and NFL/Post-NFL-Bright). Another reason is that, as mentioned earlier, we observed that more layers provide better repeatability, and the AC used more layers. The primary goal is to identify a metric that not only provides better repeatability and reproducibility but is also reliable across all datasets.

To evaluate the metrics, this study utilized two cohorts of datasets: Neutral Density Filter (NDF) experimental and the Functional and Structural OCT (FSOCT). In the NDF experiment, a set of seven optical density levels were employed, consisting of NDF with optical densities ranging from 0.1 to 0.6, in addition to NDF without any filter. It could be noted here cataract not only affect the peripapillary RNFL thickness but also signal strength [9]. The reasons we used NDF experimental data in the analysis because we wanted to test dependency of the normalized NFL reflectance in each NDF levels. The FSOCT datasets divided on two sets based on number of scans for within-visit repeatability and between visit reproducibility.

In the final analysis, we evaluated the references to test the dependencies of NFL reflectance on NDF levels, as well as the repeatability, reproducibility, and diagnostic accuracy of the FSOCT dataset. We estimated the slope against each NDF level with the NDF dataset and calculated the pooled standard deviation (pooled SD) and AROC for within-visit repeatability, between-visit reproducibility, and diagnostic accuracy, respectively.