Appearance of cysts and capillary non perfusion areas in diabetic macular edema using two different OCTA devices

Multimodal imaging in diabetic retinopathy and macular edema: An update about biomarkers

Diabetic macular edema (DME), defined as retinal thickening near, or involving the fovea caused by fluid accumulation in the retina, can lead to vision impairment and blindness in patients with diabetes. Current knowledge of retina anatomy and function and DME pathophysiology has taken great advantage of the availability of several techniques for visualizing the retina. Combining these techniques in a multimodal imaging approach to DME is recommended to improve diagnosis and to guide treatment decisions. We review the recent literature about the following retinal imaging technologies: optical coherence tomography (OCT), OCT angiography (OCTA), wide-field and ultrawide-field techniques applied to fundus photography, fluorescein angiography, and OCTA. The emphasis will be on characteristic DME features identified by these imaging technologies and their potential or established role as diagnostic, prognostic, or predictive biomarkers. The role of artificial intelligence in the assessment and interpretation of retina images is also discussed.

Glaucoma detection using non-perfused areas in OCTA

Multiple ophthalmic diseases lead to decreased capillary perfusion that can be visualized using optical coherence tomography angiography images. To quantify the decrease in perfusion, past studies have often used the vessel density, which is the percentage of vessel pixels in the image. However, this method is often not sensitive enough to detect subtle changes in early pathology. More recent methods are based on quantifying non-perfused or intercapillary areas between the vessels. These methods rely upon the accuracy of vessel segmentation, which is a challenging task and therefore a limiting factor for reliability. Intercapillary areas computed from perfusion-distance measures are less sensitive to errors in the vessel segmentation since the distance to the next vessel is only slightly changing if gaps are present in the segmentation. We present a novel method for distinguishing between glaucoma patients and healthy controls based on features computed from the probability density function of these perfusion-distance areas. The proposed approach is evaluated on different capillary plexuses and outperforms previously proposed methods that use handcrafted features for classification. Moreover the results of the proposed method are in the same range as the ones of convolutional neural networks trained on the raw input images and is therefore a computationally efficient, simple to implement and explainable alternative to deep learning-based approaches.

Association between the response of intravitreal antivascular endothelial growth factor injection and systemic factors of diabetic macular edema

BACKGROUND

This study investigated the effects of systemic factors in response to intravitreal injections in patients with macular edema due to non-proliferative diabetic retinopathy (NPDR).

METHODS

We retrospectively reviewed the medical records of patients treated with intravitreal injections for macular edema secondary to NPDR between January 2018 and January 2021. The patients were divided into three groups according to the injection response. When patients with diabetic macular edema showed 20µ or more reduction in central retinal thickness compared to baseline, they were classified as responsive group, and if not, they were classified as refractory group. The responsive group was further divided into the complete and incomplete response groups. Patients with complete disappearance of edema at seven months were classified as the complete response group, whereas those in which edema did not disappear were classified as the incomplete response group. The clinical characteristics of each group, including medical history, ophthalmic examination results, and laboratory examination results at the time of diagnosis, were analyzed.

RESULTS

Of the 112 eyes (91 patients) that satisfied the inclusion criteria, 89 (77 patients) in the responsive group and 23 (14 patients) in the refractory group were included in the analysis. The responsive group was further divided into the complete (51 eyes) and incomplete (38 eyes) response groups. The refractory group had significantly higher glycated hemoglobin levels and significantly lower estimated glomerular filtration rates than the responsive group (p = 0.026 and p = 0.012, respectively). In the multivariate logistic regression analysis, both factors were found to be significant in predicting the degree of response (all p < 0.05). No factor showed a significant difference between the incomplete and complete response groups(all p > 0.05).

CONCLUSIONS

In macular edema caused by NPDR, low glomerular filtration rates and high glycated hemoglobin levels may be used as predictors of poor response to intravitreal injection therapy. In addition to blood glucose control, education should be provided regarding the need for the continuous monitoring of renal function.

Optical Coherence Tomography Angiography as a Diagnostic Tool for Diabetic Retinopathy

Diabetic retinopathy (DR) is the most common microvascular complication of diabetes mellitus, leading to visual impairment if left untreated. This review discusses the use of optical coherence tomography angiography (OCTA) as a diagnostic tool for the early detection and management of DR. OCTA is a fast, non-invasive, non-contact test that enables the detailed visualisation of the macular microvasculature in different plexuses. OCTA offers several advantages over fundus fluorescein angiography (FFA), notably offering quantitative data. OCTA is not without limitations, including the requirement for careful interpretation of artefacts and the limited region of interest that can be captured currently. We explore how OCTA has been instrumental in detecting early microvascular changes that precede clinical signs of DR. We also discuss the application of OCTA in the diagnosis and management of various stages of DR, including non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), diabetic macular oedema (DMO), diabetic macular ischaemia (DMI), and pre-diabetes. Finally, we discuss the future role of OCTA and how it may be used to enhance the clinical outcomes of DR.

Semi-Supervised Dual Stream Segmentation Network for Fundus Lesion Segmentation

Accurate segmentation of retinal images can assist ophthalmologists to determine the degree of retinopathy and diagnose other systemic diseases. However, the structure of the retina is complex, and different anatomical structures often affect the segmentation of fundus lesions. In this paper, a new segmentation strategy called a dual stream segmentation network embedded into a conditional generative adversarial network is proposed to improve the accuracy of retinal lesion segmentation. First, a dual stream encoder is proposed to utilize the capabilities of two different networks and extract more feature information. Second, a multiple level fuse block is proposed to decode the richer and more effective features from the two different parallel encoders. Third, the proposed network is further trained in a semi-supervised adversarial manner to leverage from labeled images and unlabeled images with high confident pseudo labels, which are selected by the dual stream Bayesian segmentation network. An annotation discriminator is further proposed to reduce the negativity that prediction tends to become increasingly similar to the inaccurate predictions of unlabeled images. The proposed method is cross-validated in 384 clinical fundus fluorescein angiography images and 1040 optical coherence tomography images. Compared to state-of-the-art methods, the proposed method can achieve better segmentation of retinal capillary non-perfusion region and choroidal neovascularization.

Disentangled Representation Learning for OCTA Vessel Segmentation With Limited Training Data

Optical coherence tomography angiography (OCTA) is an imaging modality that can be used for analyzing retinal vasculature. Quantitative assessment of en face OCTA images requires accurate segmentation of the capillaries. Using deep learning approaches for this task faces two major challenges. First, acquiring sufficient manual delineations for training can take hundreds of hours. Second, OCTA images suffer from numerous contrast-related artifacts that are currently inherent to the modality and vary dramatically across scanners. We propose to solve both problems by learning a disentanglement of an anatomy component and a local contrast component from paired OCTA scans. With the contrast removed from the anatomy component, a deep learning model that takes the anatomy component as input can learn to segment vessels with a limited portion of the training images being manually labeled. Our method demonstrates state-of-the-art performance for OCTA vessel segmentation.

Cortisol awake response imbalance as an indicator of acute central serous chorioretinopathy: Relationship with choriocapillaris and choroidal features

Purpose

The purpose of the present study was to measure in central serous chorioretinopathy (CSC) the salivary cortisol awake response (CAR) delta percentage (Δ%) variation, a distinct and robust indicator of cortisol rhythm during wakefulness, commonly proposed as a marker of hypothalamic-pituitary adrenal (HPA) axis activity, whose alteration is frequently associated with several adverse health outcomes.

Methods

In the present cross-sectional observational study, salivary CAR Δ% variation was assessed in 17 adult male subjects affected by acute naïve CSC and compared to 17 matched healthy controls. Choroid vasculature metrics were assessed in the study population by measuring the subfoveal choroidal thickness (FCT) and the choroidal vascularity index (CVI) by the imaging technique of enhanced-depth imaging spectral-domain optical coherence tomography (EDI-SD-OCT). Furthermore, flow signal void area features of the choriocapillaris were evaluated in the study population using OCT angiography (OCTA).

Results

Both the control and CSC groups showed a physiological cortisol increase that occurred during the first 30 min after awaking. However, CSC adult male patients showed remarkably blunted CAR Δ% variation in comparison with controls, which might reflect a CSC-related imbalance of HPA axis activity. Statistically significant correlations were shown by Pearson’s correlation test between salivary CAR Δ% and the selected choroidal and choriocapillaris imaging biomarkers (FCT, CVI, and flow signal void area) in the study population.

Conclusion

In conclusion, alterations of the CAR Δ% increase, associated with choroidal-retinal metrics, might provide a window into the physiopathology of acute CSC, suggesting a possible common factor to explain the association between stress and CSC.

Differences in cysts characteristics and related influence on the anatomical response after dexamethasone implant in diabetic macular oedema

PURPOSE

To assess the clinical significance of suspended scattering particles in motion (SSPiM) and different cystic phenotypes in diabetic macular oedema (DME) treated with dexamethasone implant (DEX-i).

METHODS

A retrospective review of type 2 diabetic patients with DME treated with a DEX-i was conducted. Swept-source optical coherence tomography angiography (OCTA, PLEX Elite 9000) with a 3-mm volume cube was performed. Regions of interest were delineated with Fiji software (version 2.1.0/1.53.c) in the superficial vascular complex (SVC) and deep capillary plexus (DCP) at baseline, 2- and 4-months after DEX-i. SSPiM was defined as regions of variable reflectivity with a decorrelation signal. Without a detectable decorrelation signal, its counterpart was addressed as 'corpuscular,' while hyporeflective cysts were optical empty without hyperreflective material enclosed.

RESULTS

After treatment, the hyporeflective component demonstrated substantial reabsorption in the SVC (-95.4% at 2- and -84.4% at 4-months, p < 0.01 both) and DVC (-84.4%, 2-months), with a less critical decrease of the corpuscular component in the SVC (2-months: -41.9%, p = 0.001 and 4 months: -1.8%, p = 0.73), and not significant in the DVC. SSPiM did not significantly change in the SVC and DVC neither at 2- and 4-months (p > 0.05, all).

CONCLUSIONS

After a single DEX-i, the clearance of different cystic phenotypes proceeds with resorption of hyporeflective, followed by corpuscular components. SSPiM demonstrated minimal response, indicating a severe BRB breakdown that may require repeated treatment to reach a satisfactory anatomical response.

Changes in Macular Perfusion After ILUVIEN® Intravitreal Implant for Diabetic Macular Edema: An OCTA Study

INTRODUCTION

The aim of this study was to investigate changes in macular perfusion in patients affected by diabetic macular edema (DME) and treated with ILUVIEN® (fluocinolone acetonide intravitreal implant) 0.19 mg using optical coherence tomography angiography (OCTA).

METHODS

This was a retrospective cohort study that included patients aged > 18 years with type 2 non-proliferative diabetic retinopathy (DR) and DME at baseline. All patients were treated with the ILUVIEN® implant. A minimum of two 6 × 6-mm OCTA scans were required to ensure that all cases had a baseline OCTA and an OCTA performed at 4 months of follow-up. Qualitative and quantitative comparisons were performed.

RESULTS

Ten eyes from ten subjects were included in the analysis. Mean (± standard deviation) age of the study cohort was 57.1 ± 8.3 years. Mean parafoveal perfusion density (PD) at baseline was 64.1 ± 1.8% at baseline, increasing to 66.1 ± 2.9% (p = 0.013) at the 4-month follow-up visit. Mean parafoveal PD at baseline was 64.4 ± 2.1%, increasing to 65.2 ± 2.6% (p = 0.024) after 4 months. In the qualitative assessment, 60 regions (10 areas for each subject) were graded to assess changes in retinal perfusion between the baseline and follow-up visits. This assessment revealed that 24 regions (40.0%) were characterized by a qualitative increase in perfusion after treatment, while 22 (36.7%) and 14 (23.3%) regions were featured by a stability and reduction in retinal perfusion, respectively.

CONCLUSION

OCTA analysis detects improvements in macular perfusion after treatment with ILUVIEN®. This improvement in macular perfusion may be associated with corticosteroid-related beneficial effects on leukostasis.

Relationship of Choroidal Vasculature and Choriocapillaris Flow With Alterations of Salivary α-Amylase Patterns in Central Serous Chorioretinopathy

Purpose

An abnormality in choroidal vasculature is a known factor in the pathogenesis of central serous chorioretinopathy (CSC), a chorioretinal disease affecting mostly middle-aged males. The purpose of the present study was to investigate the role of the autonomic nervous system (ANS) in the pathophysiology of CSC.

Methods

This was a cross-sectional observational study in which characteristic choroidal vasculature metrics were assessed by measuring the subfoveal choroidal thickness (FCT) and the choroidal vascularity index (CVI) using the imaging technique of enhanced-depth imaging spectral-domain optical coherence tomography (EDI-SD-OCT). Furthermore, flow signal void area features were also evaluated in the study population using OCT angiography (OCTA). Diurnal patterns of salivary α-amylase (a-AMY) production, proposed as a marker of autonomic activity, were assessed in an adult male study population affected by acute naïve CSC in comparison with matched healthy controls.

Results

Results include an overall higher diurnal output of salivary a-AMY production, which is in line with the phenomenon of a sympathetic “drive” playing a role in the pathophysiology of CSC, and a flattened diurnal percentage variation in α-AMY in CSC-affected subjects. Furthermore, Pearson's coefficient test revealed statistically significant correlations between salivary α-AMY diurnal percentage variation and selected choroidal imaging biomarkers (FCT, CVI, and flow signal void area). Finally, multiple linear regression analysis identified salivary α-AMY diurnal percentage production as the sole predictor of the CVI and flow signal void area in the study population.

Conclusions

Autonomic nervous system dysregulation was highlighted in CSC patients.

Volume rendered 3D OCTA assessment of macular ischemia in patients with type 1 diabetes and without diabetic retinopathy

The aim of this study was to measure macular perfusion in patients with type 1 diabetes and no signs of diabetic retinopathy (DR) using volume rendered three-dimensional (3D) optical coherence tomography angiography (OCTA). We collected data from 35 patients with diabetes and no DR who had OCTA obtained. An additional control group of 35 eyes from 35 healthy subjects was included for comparison. OCTA volume data were processed with a previously presented algorithm in order to obtain the 3D vascular volume and 3D perfusion density. In order to weigh the contribution of different plexuses' impairment to volume rendered vascular perfusion, OCTA en face images were binarized in order to obtain two-dimensional (2D) perfusion density metrics. Mean ± SD age was 27.2 ± 10.2 years [range 19-64 years] in the diabetic group and 31.0 ± 11.4 years [range 19-61 years] in the control group (p = 0.145). The 3D vascular volume was 0.27 ± 0.05 mm³ in the diabetic group and 0.29 ± 0.04 mm³ in the control group (p = 0.020). The 3D perfusion density was 9.3 ± 1.6% and 10.3 ± 1.6% in diabetic patients and controls, respectively (p = 0.005). Using a 2D visualization, the perfusion density was lower in diabetic patients, but only at the deep vascular complex (DVC) level (38.9 ± 3.7% in diabetes and 41.0 ± 3.1% in controls, p = 0.001), while no differences were detected at the superficial capillary plexus (SCP) level (34.4 ± 3.1% and 34.3 ± 3.8% in the diabetic and healthy subjects, respectively, p = 0.899). In conclusion, eyes without signs of DR of patients with diabetes have a reduced volume rendered macular perfusion compared to control healthy eyes.

OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY ASSESSMENT OF THE DIABETIC MACULA: A Comparison Study Among Different Algorithms

PURPOSE

To assess the impact of histogram adjustments and binarization thresholding selection on quantitative measurements of diabetic macular ischemia using optical coherence tomography angiography (OCTA).

METHODS

Patients with diabetic retinopathy (DR) who had swept-source OCTA imaging obtained were enrolled. An additional group of 15 healthy control subjects was included for comparison. Previously used brightness/contrast changes and binarization thresholds were applied to original OCTA images to obtain and compare different binarized images. Qualitative and quantitative comparisons were performed.

RESULTS

Thirty patients with DR (30 eyes) were included in the analysis. Fifteen eyes displayed the presence of diabetic macular edema. Qualitative grading revealed that binarized images obtained using a global threshold had better quality compared with local or multistep thresholds. The "median" filter was most frequently graded as the histogram adjustment resulting in binarized images with best quality. In the quantitative analysis, local thresholds tended to generate higher values of measured metrics. Differences in OCTA metrics between global and local thresholds were associated with presence of diabetic macular edema and signal strength index value. In the comparison between healthy and DR eyes, differences in OCTA metrics were significantly affected by binarization threshold selection.

CONCLUSION

Quantitative OCTA parameters may be significantly influenced by strategies to quantify macular perfusion. Image quality and presence of macular edema can significantly impact OCTA-derived quantitative vascular measurements and differences between global and local binarization thresholds. These findings highlight the importance of consistent strategies to reliably generate quantitative OCTA metrics in patients with DR.

USING THREE-DIMENSIONAL OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY METRICS IMPROVES REPEATABILITY ON QUANTIFICATION OF ISCHEMIA IN EYES WITH DIABETIC MACULAR EDEMA

PURPOSE

Two-dimensional (2D) optical coherence tomography angiography (OCTA) is known to be prone to segmentation errors, especially in pathologic eyes. Therefore, our aim was to systematically compare intrasession repeatability between repeated scans for 2D and three-dimensional (3D) OCTA metrics in quantifying retinal perfusion in eyes with diabetic macular edema.

METHODS

Diabetic patients with diabetic retinopathy and diabetic macular edema who had two consecutive OCTA imaging scans obtained during the same visit were retrospectively included. A previously validated algorithm was applied to OCTA volume data to measure the 3D vascular volume and perfusion density. Optical coherence tomography angiography en face images were also processed to obtain 2D perfusion density metrics.

RESULTS

Twenty patients (20 eyes) with diabetic retinopathy and diabetic macular edema were included. The intraclass correlation coefficient ranged from 0.591 to 0.824 for 2D OCTA metrics and from 0.935 to 0.967 for 3D OCTA metrics. Therefore, compared with the 2D OCTA analysis, the intraclass correlation coefficients of the 3D OCTA analysis were higher (without overlapping of the 95% confidential intervals). Similarly, the coefficient of variation (ranging from 2.2 to 4.2 for 2D OCTA metrics and from 1.9 to 2.0 for 3D OCTA metrics) indicated that the 3D OCTA-based quantifications had the highest interscan intrasession agreements. Differences in interscan 2D OCTA metrics' values were associated with average macular volume.

CONCLUSION

Three-dimensional OCTA metrics have higher values of intrasession repeatability, as compared with 2D OCTA metrics. The latter finding seems to be related to the high rate of segmentation errors occurring in diabetic macular edema eyes.

Optical Coherence Tomography Angiography of Macular Perfusion Changes after Anti-VEGF Therapy for Diabetic Macular Edema: A Systematic Review

BACKGROUND

Diabetic macular edema (DME) is a major cause of vision loss in diabetics that is currently mainly treated by antivascular endothelial growth factor (VEGF) agents. The effect of these agents on macular perfusion (MP) is a current concern. Optical coherence tomography angiography (OCTA) is an imaging modality that allows noninvasive high-resolution retinal microvasculature imaging. Several recent studies evaluated the effect of anti-VEGF agents on the MP of DME patients using OCTA. Our aim is to provide a systematic review of these studies.

METHODS

Multiple databases were searched including PubMed, Ovid Medline, EMBASE, and Google Scholar for relevant studies published between January 2016 and November 2020 which were included in this review. Studies were compared regarding their design, the number of included patients, the machine and scanning protocol used, the inclusion and exclusion criteria, the number of injections given, the type of anti-VEGF agent used, the outcome measures assessed, and the effect of injections on different MP parameters.

RESULTS

A total of 16 studies were included. The studies assessed various OCTA parameters that define MP including the foveal avascular zone area and superficial and deep vascular density and yielded conflicting results. Seven studies showed stable or improved MP following treatment, while 7 studies showed worsening MP following treatment, and 2 studies showed inconclusive results. This could have been due to differences in study design, inclusion criteria, type of anti-VEGF agents used, treatment duration, and methods of image analysis and vascular density quantification. All identified studies were noncomparative case series, and 14 of them (87.5%) used the RTVue XR Avanti OCTA machine. Only one study compared OCTA to fluorescein angiography findings.

CONCLUSION

Analysis of MP changes following VEGF inhibition for DME could benefit from a unified scanning protocol and analysis approach that uses similar study designs to eliminate potential sources of bias. This may provide more definitive conclusions regarding the effect of treatment on MP.

A Swept source optical coherence tomography angiography study: Imaging artifacts and comparison of non-perfusion areas with fluorescein angiography in diabetic macular edema

Purpose

Swept Source Optical coherence tomography angiography (SS-OCTA) is a novel technique to visualize perfusion and vascular changes like ischemia in patients with diabetic retinopathy. The aim of this study was to compare non-perfusion areas on conventional fluorescein angiography (FA) with those on SS-OCTA using detailed manual annotation in patients with diabetic macular edema (DME) and to evaluate possible artifacts caused by DME on SS-OCTA.

Methods

27 eyes of 21 patients with DME were analyzed in this prospective, cross-sectional study; on all, standard ophthalmological examination, SS-OCTA and FA imaging were performed. Early-phase FA and SS-OCTA images were analyzed for capillary dropout and foveal avascular zone (FAZ) was measured on both modalities. Artifacts in SS-OCTA imaging caused by DME were marked and analyzed.

Results

The mean age of the patients was 62.6 ± 11.5 years. On FA the mean size of the annotated non-perfusion areas was 0.14 ± 0.31 mm² whereas the mean size in SS-OCTA was 0.04 ± 0.13 mm²; areas marked on FA were statistically significantly larger than on SS-OCTA (p<0.01). Mean size of FAZs was similar between FA and OCTA images. (p = 0.91). Seven eyes (25.9 percent) showed imaging artifacts due to DME in SS-OCTA.

Conclusion

SS-OCTA is a valid tool to analyze capillary perfusion status of patients with DME, although areas of non-perfusion were measured smaller than in conventional FA. More non-perfusion areas were found on SS-OCTA images. FAZ measurements were similar using the two modalities. However, SS-OCTA is prone to artifacts and therefore requires reviewing of imaging results: up to 25 percent of the analyzed eyes showed artifacts on OCTA, which occurred in the areas of diabetic macular edema and did not correspond to capillary drop out.

Vessel Density Changes on Optical Coherence Tomography Angiography after Vascular Endothelial Growth Factor Inhibitor Treatment for Diabetic Macular Edema

Objectives

To evaluate the changes in macular vessel density after treatment with vascular endothelial growth factor (VEGF) inhibitors in center-involving diabetic macular edema (DME) and to compare these changes between anatomical responders and non-responders.

Materials and Methods

This retrospective study included 22 eyes with center-involving DME. All eyes had 3 consecutive administrations of VEGF inhibitors. Optical coherence tomography (OCT) and OCT angiography (OCTA) of the macula with manual adjustment of segmentation lines were performed at baseline and after treatment. Vessel density in the central and parafoveal regions of the superficial and deep capillary plexus (SCP/DCP) were measured at baseline and after treatment. Vessel density and changes therein were compared between anatomical responders and non-responders as defined by changes in central subfield thickness (CST).

Results

Overall, there were no significant differences in vessel density in the central and parafoveal regions of the SCP and DCP after treatment compared to baseline. After categorization by anatomical response, 12 eyes were responders (CST decreased by 173.7±47.7 μm) and 10 eyes were non-responders (CST increased by 20.8±38.9 μm) (p<0.0001). There were no corresponding significant differences between responders and non-responders in SCP and DCP vessel density or changes therein after treatment.

Conclusion

There were no significant changes in macular vessel density after the early stages of VEGF inhibitor treatment for DME, and there was no relationship with the anatomical response. The effect of VEGF inhibitors in DME treatment may not be related to increasing vessel density.

Quantification of diabetic macular ischemia using novel three-dimensional optical coherence tomography angiography metrics

We applied three-dimensional (3D) analysis to optical coherence tomography angiography (OCTA) to measure macular ischemia in eyes affected by non-proliferative diabetic retinopathy (DR). A previously validated algorithm was applied to OCTA data in order to obtain 3D visualization of the retinal vasculature. Successively, a global thresholding algorithm was applied and two novel quantitative metrics were introduced: 3D vascular volume and 3D perfusion density. Two-dimensional (2D) OCTA metrics were also obtained with different binarization thresholds for comparison. Of the 30 patients included, 15 were diagnosed with DR and 15 were controls. The 3D vascular volume and 3D perfusion density were reduced in DR eyes (P < .0001). The 2D variables also significantly differ between groups. The 3D perfusion density had the highest area under the receiver operating characteristic curve (0.964) among tested variables. Assessing quantitative perfusion using 3D analysis is reliable and promising, and with an elevated diagnostic efficacy in identifying DR eyes.