Spectral-domain Optical Coherence Tomography B-scans Research Articles

Optical coherence tomography choroidal enhancement using generative deep learning

Spectral-domain optical coherence tomography (SDOCT) is the gold standard of imaging the eye in clinics. Penetration depth with such devices is, however, limited and visualization of the choroid, which is essential for diagnosing chorioretinal disease, remains limited. Whereas swept-source OCT (SSOCT) devices allow for visualization of the choroid these instruments are expensive and availability in praxis is limited. We present an artificial intelligence (AI)-based solution to enhance the visualization of the choroid in OCT scans and allow for quantitative measurements of choroidal metrics using generative deep learning (DL). Synthetically enhanced SDOCT B-scans with improved choroidal visibility were generated, leveraging matching images to learn deep anatomical features during the training. Using a single-center tertiary eye care institution cohort comprising a total of 362 SDOCT-SSOCT paired subjects, we trained our model with 150,784 images from 410 healthy, 192 glaucoma, and 133 diabetic retinopathy eyes. An independent external test dataset of 37,376 images from 146 eyes was deployed to assess the authenticity and quality of the synthetically enhanced SDOCT images. Experts’ ability to differentiate real versus synthetic images was poor (47.5% accuracy). Measurements of choroidal thickness, area, volume, and vascularity index, from the reference SSOCT and synthetically enhanced SDOCT, showed high Pearson’s correlations of 0.97 [95% CI: 0.96–0.98], 0.97 [0.95–0.98], 0.95 [0.92–0.98], and 0.87 [0.83–0.91], with intra-class correlation values of 0.99 [0.98–0.99], 0.98 [0.98–0.99], and 0.95 [0.96–0.98], 0.93 [0.91–0.95], respectively. Thus, our DL generative model successfully generated realistic enhanced SDOCT data that is indistinguishable from SSOCT images providing improved visualization of the choroid. This technology enabled accurate measurements of choroidal metrics previously limited by the imaging depth constraints of SDOCT. The findings open new possibilities for utilizing affordable SDOCT devices in studying the choroid in both healthy and pathological conditions.

Deep learning-based algorithm for the detection of idiopathic full thickness macular holes in spectral domain optical coherence tomography

BackgroundAutomated identification of spectral domain optical coherence tomography (SD-OCT) features can improve retina clinic workflow efficiency as they are able to detect pathologic findings. The purpose of this study was to test a deep learning (DL)-based algorithm for the identification of Idiopathic Full Thickness Macular Hole (IFTMH) features and stages of severity in SD-OCT B-scans.MethodsIn this cross-sectional study, subjects solely diagnosed with either IFTMH or Posterior Vitreous Detachment (PVD) were identified excluding secondary causes of macular holes, any concurrent maculopathies, or incomplete records. SD-OCT scans (512 × 128) from all subjects were acquired with CIRRUS™ HD-OCT (ZEISS, Dublin, CA) and reviewed for quality. In order to establish a ground truth classification, each SD-OCT B-scan was labeled by two trained graders and adjudicated by a retina specialist when applicable. Two test sets were built based on different gold-standard classification methods. The sensitivity, specificity and accuracy of the algorithm to identify IFTMH features in SD-OCT B-scans were determined. Spearman’s correlation was run to examine if the algorithm’s probability score was associated with the severity stages of IFTMH.ResultsSix hundred and one SD-OCT cube scans from 601 subjects (299 with IFTMH and 302 with PVD) were used. A total of 76,928 individual SD-OCT B-scans were labeled gradable by the algorithm and yielded an accuracy of 88.5% (test set 1, 33,024 B-scans) and 91.4% (test set 2, 43,904 B-scans) in identifying SD-OCT features of IFTMHs. A Spearman’s correlation coefficient of 0.15 was achieved between the algorithm’s probability score and the stages of the 299 (47 [15.7%] stage 2, 56 [18.7%] stage 3 and 196 [65.6%] stage 4) IFTMHs cubes studied.ConclusionsThe DL-based algorithm was able to accurately detect IFTMHs features on individual SD-OCT B-scans in both test sets. However, there was a low correlation between the algorithm’s probability score and IFTMH severity stages. The algorithm may serve as a clinical decision support tool that assists with the identification of IFTMHs. Further training is necessary for the algorithm to identify stages of IFTMHs.

Retinal Pigment Epithelial Detachments Devoid of Retinal Pigment Epithelium.

To describe two patients with chronic central serous chorioretinopathy (CSC) showing what appeared to be retinal pigment epithelium detachments (PED) lacking imaging findings consistent with retinal pigment epithelium (RPE) over the elevation. The patients underwent comprehensive ophthalmic examination, including multicolor fundus photography, fundus autofluorescence (FAF), and spectral-domain optical coherence tomography (SD-OCT). A 70-year-old male and a 58-year-old male, both diagnosed with chronic CSC, showed PED-like lesions that were hypoautofluorescent, suggesting an absence of RPE. SD-OCT B-scans showed serous, dome-shaped elevations composed of a narrow, mildly hyperreflective band (9-10 μm thick) that demonstrated hypertransmission of light. The material that constituted the elevation was contiguous with the outer portion of the RPE band at the lesion borders. Based on the multimodal imaging findings we hypothesize that these elevations of the retina have lost their overlying RPE. A thin layer of material that could represent a residual layer of basal laminar deposit produced by the RPE remains overlying the detachments, possibly accounting for their dome shape and structural stability.

Correction of Retinal Nerve Fiber Layer Thickness Measurement on Spectral-Domain Optical Coherence Tomographic Images Using U-net Architecture.

In this study, an algorithm based on deep learning was presented to reduce the retinal nerve fiber layer (RNFL) segmentation errors in spectral domain optical coherence tomography (SD-OCT) scans using ophthalmologists' manual segmentation as a reference standard. In this study, we developed an image segmentation network based on deep learning to automatically identify the RNFL thickness from B-scans obtained with SD-OCT. The scans were collected from Farabi Eye Hospital (500 B-scans were used for training, while 50 were used for testing). To remove the speckle noise from the images, preprocessing was applied before training, and postprocessing was performed to fill any discontinuities that might exist. Afterward, output masks were analyzed for their average thickness. Finally, the calculation of mean absolute error between predicted and ground truth RNFL thickness was performed. Based on the testing database, SD-OCT segmentation had an average dice similarity coefficient of 0.91, and thickness estimation had a mean absolute error of 2.23 2.1 μm. As compared to conventional OCT software algorithms, deep learning predictions were better correlated with the best available estimate during the test period (r2 = 0.99 vs r2 = 0.88, respectively; P 0.001). Our experimental results demonstrate effective and precise segmentation of the RNFL layer with the coefficient of 0.91 and reliable thickness prediction with MAE 2.23 2.1 μm in SD-OCT B-scans. Performance is comparable with human annotation of the RNFL layer and other algorithms according to the correlation coefficient of 0.99 and 0.88, respectively, while artifacts and errors are evident.

Automatic segmentation of hyperreflective dots via focal priors and visual saliency.

Hyperreflective dots (HRDs) can be observed in spectral domain optical coherence tomography (SD-OCT), which can provide a sensitive marker in the treatment decision process. Quantitative analyses of HRDs are the key to make appropriate decisions on observation, treatment, and retreatment. The purpose of this study is to automatically and accurately segment HRDs in SD-OCT B-scans with diabetic retinopathy (DR). The authors propose an automatic segmentation algorithm of HRDs via focal priors and visual saliency. The algorithm is divided into three stages: segmentation of retinal layers, calculation of the multiscale local contrast saliency map, and adaptive threshold segmentation. First, a method based on improved graph search is used to segment retinal layers to obtain the region of interest (ROI) and the reflectivity estimation of the retinal pigment epithelium (RPE) layer; then, the multiscale local contrast saliency map is obtained by using a local contrast measure, which measures the dissimilarity between the current pixels and corresponding neighborhoods; finally, an adaptive threshold is applied to segment HRDs. Experimental results on 20 SD-OCT B-scans demonstrate that our method is effective for HRDs segmentation. The average dice similarity coefficient (DSC) and detection accuracy are 71.12% and 85.07%, respectively. The proposed method can accurately segment HRDs in SD-OCT B-scans with DR and outperforms current state-of-the-art methods. Our method can provide reliable HRDs segmentation to assist ophthalmologists in clinical diagnosis, treatment, disease monitoring, andprogression.

Macular Telangiectasia type 2: multimodal assessment of retinal function and microstructure.

To assess the impact of neurodegenerative morphologic alterations due to macular telangiectasia type 2 (MacTel) on microperimetry (MP) and multifocal electroretinography (mfERG). Thirty-five eyes of 18 patients with MacTel were examined using spectral domain optical coherence tomography (SD-OCT), fundus autofluorescence (FAF), mfERG and MP. Software was used to match SD-OCT B-scans with the corresponding retinal sensitivity map and multifocal electroretinograms (mfERGs), thus enabling direct structure/function correlation. Loss of the ellipsoid zone (EZ) had the strongest negative association with retinal sensitivity (16.77 dB versus 4.58 dB, adj. p < 0.001) of all parameters examined, and a limited negative effect on mfERGs (0.32 SD versus -1.97 SD adj. p = 0.121). Ellipsoid zone (EZ) irregularity was associated with reduced MP values but preserved mfERGs. There was a significant association between areas of inner retinal hyporeflectivity and loss of MP sensitivity (adj. p < 0.001) but the reduction in sensitivity was less than in locations with EZ loss. Areas of mfERG abnormality showed similar sensitivity loss with either inner retinal hyporeflectivity or EZ loss (adj. p = 0.063). In areas with EZ loss alone, preservation of the external limiting membrane (ELM) was associated with higher MP values than in areas with additional ELM loss; the integrity of the ELM alone was not associated with changes either in MP or mfERG. Increased FAF was observed in 51% of eyes, mixed/reduced FAF in 40%, and no abnormality was detected in 9% of eyes. The data suggest both MP and mfERG to be useful non-invasive modalities for detecting localised macular dysfunction in MacTel. The findings suggest a different sensitivity of the two modalities to inner and outer retinal changes in macular function and are therefore complementary.

Deep Learning-Based Retinal Nerve Fiber Layer Thickness Measurement of Murine Eyes.

PurposeTo design a robust and automated estimation method for measuring the retinal nerve fiber layer (RNFL) thickness using spectral domain optical coherence tomography (SD-OCT).MethodsWe developed a deep learning–based image segmentation network for automated segmentation of the RNFL in SD-OCT B-scans of mouse eyes. In total, 5500 SD-OCT B-scans (5200 B-scans were used as training data with the remaining 300 B-scans used as testing data) were used to develop this segmentation network. Postprocessing operations were then applied on the segmentation results to fill any discontinuities or remove any speckles in the RNFL. Subsequently, a three-dimensional retina thickness map was generated by z-stacking 100 segmentation processed thickness B-scan images together. Finally, the average absolute difference between algorithm predicted RNFL thickness compared to the ground truth manual human segmentation was calculated.ResultsThe proposed method achieves an average dice similarity coefficient of 0.929 in the SD-OCT segmentation task and an average absolute difference of 0.0009 mm in thickness estimation task on the basis of the testing dataset. We also evaluated our segmentation algorithm on another biological dataset with SD-OCT volumes for RNFL thickness after the optic nerve crush injury. Results were shown to be comparable between the predicted and manually measured retina thickness values.ConclusionsExperimental results demonstrate that our automated segmentation algorithm reliably predicts the RNFL thickness in SD-OCT volumes of mouse eyes compared to laborious and more subjective manual SD-OCT RNFL segmentation.Translational RelevanceAutomated segmentation using a deep learning–based algorithm for murine eye OCT effectively and rapidly produced nerve fiber layer thicknesses comparable to manual segmentation.

Agreement Between Trend-Based and Qualitative Analysis of the Retinal Nerve Fiber Layer Thickness for Glaucoma Progression on Spectral-Domain Optical Coherence Tomography.

IntroductionTo evaluate the agreement between trend-based analysis and qualitative assessment of the retinal nerve fiber layer (RNFL) thickness for glaucomatous progression on spectral-domain optical coherence tomography (SDOCT).MethodsRetrospective review of 190 eyes from 103 patients with glaucoma or suspected glaucoma that underwent SDOCT imaging during four consecutive clinic visits. Trend-based progression was characterized by a significantly negative slope. Progression by qualitative analysis was determined by review of raw SDOCT B-scans.ResultsThe slope was significantly greater in those with progression than without progression for both trend-based and qualitative analysis (p < 0.001). However, the qualitative grading classified a significantly greater proportion of eyes as progressing compared to trend-based analysis in both the superotemporal (ST) (23.2% vs. 10.5%, p = 0.001) and inferotemporal (IT) RNFL (27.4% vs 8.4%, p < 0.001). The trend-based and qualitative classifications of progression showed poor agreement in both the ST (kappa = 0.0135) and IT RNFL (kappa = 0.1222). The agreement between trend-based and qualitative analysis was lower for eyes with artifacts (ST = 58.11%; IT = 68.7%) than those without artifacts (ST = 80.2%; IT = 74.8%). Moreover, among eyes with artifacts, there was no significant difference in slope between those qualitatively categorized as progressing versus not progressing (p > 0.05).ConclusionsPoor agreement was found between a trend-based and qualitative analysis of change in RNFL on SDOCT. Careful qualitative review of SDOCT imaging may identify specific areas of glaucoma progression not captured by trend-based methods, especially in the presence of artifacts. Such an approach may also prove useful for detecting glaucoma progression in a clinical setting when there are few data points available.

Accuracy of a Machine-Learning Algorithm for Detecting and Classifying Choroidal Neovascularization on Spectral-Domain Optical Coherence Tomography

Background: To evaluate the performance of a machine-learning (ML) algorithm to detect and classify choroidal neovascularization (CNV), secondary to age-related macular degeneration (AMD) on spectral-domain optical coherence tomography (SD-OCT) images. Methods: Baseline fluorescein angiography (FA) and SD-OCT images from 1037 treatment-naive study eyes and 531 fellow eyes, without advanced AMD from the phase 3 HARBOR trial (NCT00891735), were used to develop, train, and cross-validate an ML pipeline combining deep-learning–based segmentation of SD-OCT B-scans and CNV classification, based on features derived from the segmentations, in a five-fold setting. FA classification of the CNV phenotypes from HARBOR was used for generating the ground truth for model development. SD-OCT scans from the phase 2 AVENUE trial (NCT02484690) were used to externally validate the ML model. Results: The ML algorithm discriminated CNV absence from CNV presence, with a very high accuracy (area under the receiver operating characteristic [AUROC] = 0.99), and classified occult versus predominantly classic CNV types, per FA assessment, with a high accuracy (AUROC = 0.91) on HARBOR SD-OCT images. Minimally classic CNV was discriminated with significantly lower performance. Occult and predominantly classic CNV types could be discriminated with AUROC = 0.88 on baseline SD-OCT images of 165 study eyes, with CNV from AVENUE. Conclusions: Our ML model was able to detect CNV presence and CNV subtypes on SD-OCT images with high accuracy in patients with neovascular AMD.

Predicting Age From Optical Coherence Tomography Scans With Deep Learning.

PurposeTo assess whether age can be predicted from deep learning analysis of peripapillary spectral-domain optical coherence tomography (SD-OCT) B-scans and to determine the importance of specific retinal areas on the predictions.MethodsDeep learning (DL) convolutional neural networks were developed to predict chronological age in healthy subjects using peripapillary SD-OCT B-scan images. Models were built using the whole B-scan, as well as using specific regions through image ablation. Cross-validation was used for training and testing the model. Mean absolute error (MAE) and correlations between predicted and observed age were used to evaluate model performance.ResultsA total of 7271 images from 542 eyes of 278 healthy subjects were included. DL predictions of age using the whole B-scan were strongly correlated with chronological age (MAE = 5.82 years; r = 0.860, P < 0.001). The model also accurately discriminated between the lowest and highest tertiles of age, with an area under the receiver operating characteristic curve of 0.962. In general, class activation maps tended to show a diffuse pattern of activation throughout the scan image. For specific structures of the B-scan, the layers with the strongest correlations with chronological age were the choroid and vitreous (both r = 0.736), whereas retinal nerve fiber layer had the lowest correlation (r = 0.492).ConclusionsA DL algorithm was able to accurately predict age from whole peripapillary SD-OCT B-scans.Translational RelevanceDL models applied to SD-OCT scans suggest that aging appears to affect several layers in the posterior eye segment.

Quantification of Translaminar Pressure Gradient (TLPG) With Continuous Wireless Telemetry in Nonhuman Primates (NHPs).

PurposeRecent retrospective clinical and animal studies suggest that cerebrospinal fluid pressure (CSFP) is important in glaucoma pathogenesis. Intraocular pressure (IOP) and CSFP are the driving components of translaminar pressure (TLP = IOP – CSFP), which acts across the lamina cribrosa (LC) thickness to create the translaminar pressure gradient (TLPG = TLP/LC thickness).MethodsWe developed an implantable wireless telemetry system based on a small piezoelectric sensor with low temporal drift. IOP, measured in the anterior chamber, and intracranial pressure (ICP), measured in the brain parenchyma (as a surrogate for CSFP) were measured at 200 Hz in three male rhesus macaques (nonhuman primates, NHPs) on a 10% duty cycle (15 seconds of every 150-second period). Three-dimensional LC thickness was autosegmented as the mean thickness of the visible hyperreflective band in 48 radial spectral-domain optical coherence tomography b-scans centered on the optic nerve head.ResultsResults indicated the rank order of IOP, ICP, TLP, and TLPG for waking, sleeping, and 24-hour periods averaged across all days. NHP 150110 had the highest IOP and ICP in all periods; however, it had the lowest TLPG in all periods due to its relatively thick LC. The other two NHPs showed similar shifts in the rank order of possible glaucoma risk factors.ConclusionsIOP is the only modifiable and readily measurable pressure-based risk factor for glaucoma. However, other potential risk factors such as ICP, TLP, and TLPG, as well as their rank-order patterns, differed compared to IOP across subjects, demonstrating that a comprehensive view of relevant risk factors is warranted.Translational RelevanceFuture studies should consider including CSFP, TLP, and TLPG in addition to IOP as potential risk factors when assessing eye-specific glaucoma susceptibility.

Spectral Domain Optical Coherence Tomography in Awake Rabbits Allows Identification of the Visual Streak, a Comparison with Histology.

PurposeTo evaluate visual streak (VS) identification on spectral-domain optical coherence tomography (SD-OCT) scans in awake rabbits. To report thickness measurements in the VS and adjacent retina on OCT B-scans and histologic sections and to assess inter-method bias, precision and repeatability between OCT and histology.MethodsVertical SD-OCT B-scan images through the optic nerve head and VS were acquired from 16 awake, ophthalmologically healthy experimental rabbits. Scans were acquired from both eyes, which were later enucleated and processed for light microscopy. Inner retina, inner nuclear layer, outer nuclear layer, outer retina (OR) and photoreceptor outer segment (PROS) thickness were measured on OCT images and digitalized microscopy slides in- and outside of the VS, and compared using linear mixed effects models.ResultsBoth SD-OCT and histology allowed retinal layer identification and measurement. On OCT, OR and PROS were thickest in the central VS and thinnest outside the VS. Histology mirrored OCT results for central outer retinal layers but shows discrepancies for other layers likely because of postmortem processing artifacts. The method comparison demonstrated better repeatability for OCT measurements compared with histology.ConclusionsIncreased OR and PROS thickness compared with the adjacent retina allowed identification of the VS on SD-OCT in awake rabbits. OCT allows measurements devoid of processing artifacts in contrast to histology.Translational RelevanceSD-OCT is possible in awake rabbits. Easy and reliable identification of the VS may facilitate the positioning and use of rabbits as model species in human macular and generalized retinal disease research

Peripapillary Choroidal Vascularity Index and Microstructure of Parapapillary Atrophy.

To investigate the association between the microstructure of β-zone parapapillary atrophy (βPPA) and choroidal vascularity index (CVI) determined by spectral-domain optical coherence tomography (SD-OCT) in glaucomatous eyes. A total of 160 eyes of 160 primary open-angle glaucoma patients with βPPA were included. Total choroidal area (TCA), luminal area (LA), and CVI were measured at a 3.5-mm distance from the Bruch's membrane (BM) opening center by image binarization of SD-OCT B-scans. The widths of βPPA with BM (βPPA+BM) and without BM (βPPA-BM), and juxtapapillary choroidal thickness (JPCT) were measured on six radial SD-OCT images. OCT angiography-derived parapapillary deep-layer microvasculature dropout (MvD_P) was also derived. In the multivariate regression analysis, larger βPPA+BM was significantly associated with smaller TCA and smaller LA (P < 0.05, respectively), but not with CVI and JPCT (P > 0.05, respectively). Meanwhile, βPPA-BM was not significantly associated with TCA, LA, CVI, or JPCT in the multivariate regression analysis (P > 0.05). Despite significant relationship between the choroidal thinning and larger βPPA+BM, choroidal vascularity was not associated with the βPPA+BM width. These findings suggest that the presumed common pathogenic mechanism between RPE atrophy and peripapillary choroidal thinning may not be mediated by the impaired choroidal perfusion in glaucomatous eyes. Future studies on the mechanisms in explaining the relationship between the atrophy of retinal pigment epithelium (RPE) and choroid in glaucoma are needed.

Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm

In this study we developed a deep learning (DL) algorithm that detects errors in retinal never fibre layer (RNFL) segmentation on spectral-domain optical coherence tomography (SDOCT) B-scans using human grades as the reference standard. A dataset of 25,250 SDOCT B-scans reviewed for segmentation errors by human graders was randomly divided into validation plus training (50%) and test (50%) sets. The performance of the DL algorithm was evaluated in the test sample by outputting a probability of having a segmentation error for each B-scan. The ability of the algorithm to detect segmentation errors was evaluated with the area under the receiver operating characteristic (ROC) curve. Mean DL probabilities of segmentation error in the test sample were 0.90 ± 0.17 vs. 0.12 ± 0.22 (P < 0.001) for scans with and without segmentation errors, respectively. The DL algorithm had an area under the ROC curve of 0.979 (95% CI: 0.974 to 0.984) and an overall accuracy of 92.4%. For the B-scans with severe segmentation errors in the test sample, the DL algorithm was 98.9% sensitive. This algorithm can help clinicians and researchers review images for artifacts in SDOCT tests in a timely manner and avoid inaccurate diagnostic interpretations.

Retinal Angiomatous Proliferation: Multimodal Imaging Characteristics and Follow-up with Eye-Tracked Spectral Domain Optical Coherence Tomography of Precursor Lesions.

Objectives:To present the multimodal imaging characteristics of precursor retinal angiomatous proliferation (RAP) lesions and follow-up results with eye-tracked spectral-domain optical coherence tomography (SD-OCT).Materials and Methods:Six eyes of 6 patients aged 77.5±5.9 years diagnosed with precursor RAP lesion were included in this prospective observational case series. Best corrected visual acuity (BCVA) measurement and complete ophthalmologic examination were performed for all subjects, as well as fundus photography (FP), fundus autofluorescence (FAF), SD-OCT, fluorescein angiography (FA), indocyanine green angiography (ICGA), optical coherence tomography angiography (OCTA), and their long-term follow-up results are presented.Results:The mean BCVA was 0.8±0.16 (0.6-1) Snellen and the mean follow-up was 26.3±14.8 months. Images of the precursor RAP lesions demonstrated no specific findings on FP and FAF, showed focal hypofluorescent foci with no leakage on FA and IGA, and appeared as extrafoveal small, round, well-defined, hyperreflective foci typically located in the outer retinal layers on SD-OCT B-scans with high sampling density. OCTA demonstrated the precursor lesions as the deep capillary plexus abnormalities in 3 eyes. Two eyes progressed to stage 1 RAP during the follow-up period.Conclusion:This study defined the diagnostic characteristics and clinical course of precursor RAP lesions. Our findings highlight the importance of B-scans with high sampling density for the diagnosis of precursor lesions and using eye-tracking mode SD-OCT during follow-up.

Focal Lamina Cribrosa Defect in Myopic Eyes With Nonprogressive Glaucomatous Visual Field Defect

To investigate focal lamina cribrosa (LC) defect that spatially correspond to the nonprogressive glaucomatous visual field defect (VFD) in myopic subjects. Case-control study. We included 159 myopic eyes with glaucomatous VFD under treatment and followed up for 7 years. Serial enhanced-depth imaging spectral-domain optical coherence tomography B-scans of the optic discs were acquired at the end of the follow-up and reviewed for the LC defect. Nonprogressive VFD was defined as having ≤1 progressing point of Humphrey visual field, with a slope calculated using pointwise linear regression worse than -1.0 dB/year at P < .01. Eyes were classified as having either progressive or nonprogressive VFD, and associating factors were evaluated. Sixty-four subjects (40.3%) exhibited nonprogressive VFD with mean deviation change -0.06 ± 0.22 dB/year. Multivariate logistic regression analysis revealed that presence of LC defect was significantly associated with nonprogressive VFD (odds ratio, 3.96; P= .002). The location of LC defect corresponded spatially to the location of VFD. Nonprogressive eyes with LC defect exhibited lower baseline intraocular pressure (IOP) (16.6mm Hg vs 21.0mm Hg, P= .0030) and smaller percentage of IOP change (12.9% vs 30.5%, P < .0001) than those without LC defect, but greater myopic optic disc deformation (10.1 degrees vs 1.2 degrees in torsion angle, P < .0001). When the eyes with LC defect had higher baseline IOP, they exhibited progressive VFD. In myopic eyes, there are specific patters of LC defect that are suggested to be associated with nonprogressive glaucomatous VFD.

Differentiating Occult Branch Retinal Artery Occlusion from Primary Open-angle Glaucoma.

Clinical findings in occult branch retinal artery occlusion (BRAO) can mimic those of primary open-angle glaucoma (POAG). Because management of these conditions substantially differs, accurate diagnosis is crucial. Our comparative analysis indicates that specific macular thickness variables reliably differentiate these conditions and that macular scanning may enhance routine glaucoma evaluation. The aim of this study was to identify clinical factors that reliably and efficiently identify occult BRAO masquerading as POAG. All subjects had comprehensive eye examinations including measurements of retinal nerve fiber layer and macular thickness (MT) using spectral-domain optical coherence tomography (SD-OCT). All subjects were asymptomatic for previous acute vision loss episodes, had optic nerve appearances suggestive of glaucoma, and exhibited SD-OCT retinal nerve fiber layer thinning with corresponding visual field loss. Macular thickness scans were divided into 64 individual thickness blocks with thin MT blocks defined by the lower 99% confidence interval from a group of normal eyes. We defined BRAO by the presence of regional inner retinal thinning with lack of inner-layer stratification on macular SD-OCT b-scan images that spatially corresponded with arteriolar distribution and visual field loss location. Primary open-angle glaucoma eyes were selected to match the BRAO eyes by age and disease severity. Pairwise and receiver operating characteristic curve analyses were used to compare occult BRAO and POAG eyes. Compared with POAG (n = 52), occult BRAO eyes (n = 11) demonstrated lower cup-disc ratio, greater intereye and intraeye (superior vs. inferior) MT asymmetry, and higher frequency of thin MT blocks (<200 μm). Area under the receiver operating characteristic curve (AUC) for differentiating these conditions was highest for intraeye MT asymmetry (AUC = 0.990 [95% confidence interval, 0.925 to 1.000]) and number of thin MT blocks (AUC = 0.993 [95% confidence interval, 0.929 to 1.000]). Macular thickness parameters provided accurate and efficient diagnostic capability in this study. Considering the clinical implications of inaccurate diagnosis, macular scanning may be important in baseline glaucoma evaluation.

Effect of Cataract Grade according to Wide-Field Fundus Images on Measurement of Macular Thickness in Cataract Patients.

PurposeTo investigate the effects of cataract grade based on wide-field fundus imaging on macular thickness measured by spectral domain optical coherence tomography (SD-OCT) and its signal-to-noise ratio (SNR).MethodsTwo hundred cataract patients (200 eyes) with preoperative measurements by wide-field fundus imaging and macular SD-OCT were enrolled. Cataract severity was graded from 1 to 4 according to the degree of macular obscuring by cataract artifact in fundus photo images. Cataract grade based on wide-field fundus image, the Lens Opacity Classification System III, macular thickness, and SD-OCT SNR were compared. All SD-OCT B-scan images were evaluated to detect errors in retinal layer segmentation.ResultsCataract grade based on wide-field fundus imaging was positively correlated with grade of posterior subcapsular cataracts (rho = 0.486, p < 0.001), but not with nuclear opalescence or cortical cataract using the Lens Opacity Classification System III. Cataract grade was negatively correlated with total macular thickness (rho = −0.509, p < 0.001) and SD-OCT SNR (rho = −0.568, p < 0.001). SD-OCT SNR was positively correlated with total macular thickness (rho = 0.571, p < 0.001). Of 200 eyes, 97 (48.5%) had segmentation errors on SD-OCT. As cataract grade increased and SD-OCT SNR decreased, the percentage of eyes with segmentation errors on SD-OCT increased. All measurements of macular thickness in eyes without segmentation errors were significantly greater than those of eyes with segmentation errors.ConclusionsPosterior subcapsular cataracts had profound effects on cataract grade based on wide-field fundus imaging. As cataract grade based on wide-field fundus image increased, macular thickness tended to be underestimated due to segmentation errors in SD-OCT images. Segmentation errors in SD-OCT should be considered when evaluating macular thickness in eyes with cataracts.

Multiple Temporal Lamina Cribrosa Defects in Myopic Eyes with Glaucoma and Their Association with Visual Field Defects

To investigate characteristics of lamina cribrosa (LC) defects in myopic eyes with open-angle glaucoma (OAG) using spectral-domain (SD) optical coherence tomography (OCT). Cross-sectional study. One hundred thirty-three eyes with OAG and 83 eyes without OAG, with axial length of 24 mm or more. Serial enhanced depth imaging SD OCT B-scans of the optic disc were acquired and reviewed for LC defects (diameter, ≥100 μm) and large pores (diameter, 60-100 μm). The numbers and locations of LC defects and large pores in each eye were assessed. In eyes with OAG, factors related to the number of LC defects were evaluated, as well as the association between the locations of LC defects and visual field (VF) defects (e.g.,paracentral scotoma [PCS] and superior or inferior hemifield defects). Numbers and locations of LC defects and large pores. In myopic eyes with and without OAG, the average numbers of LC defects were 3.8 and 0.8 and numbers of large pores were 1.9 and 1.6, respectively. In both groups, LC defects and large pores were located predominantly at the temporal periphery. Among eyes with OAG, the number of LC defects was relatively high in the eyes with greater optic disc tilt angle and worse mean deviation of the VF (both P < 0.001). The number of temporal LC defects and tilt angle were associated with presence of PCS, whereas the number of inferior and superior LC defects and torsion direction were associated with presence of superior and inferior VF defects. Myopic eyes with OAG exhibited LC defects and large pores at similar locations as those without OAG, but in greater numbers, suggesting that these focal alternations of the LC in myopic eyes may evolve into larger defects when glaucoma develops in the eye. The number of LC defects, which was related to the optic disc tilt angle, was associated significantly with glaucomatous VF defects in both severity and location. This suggests that myopia may influence glaucomatous VF defects through optic disc tilt by way of an increased number of LC defects at the temporal periphery.

SPECTRAL DOMAIN-OPTICAL COHERENCE TOMOGRAPHY IMAGE CONTRAST AND BACKGROUND COLOR SETTINGS INFLUENCE IDENTIFICATION OF RETINAL STRUCTURES.

To evaluate image contrast and color setting on assessment of retinal structures and morphology in spectral-domain optical coherence tomography. Two hundred and forty-eight Spectralis spectral-domain optical coherence tomography B-scans of 62 patients were analyzed by 4 readers. B-scans were extracted in 4 settings: W + N = white background with black image at normal contrast 9; W + H = white background with black image at maximum contrast 16; B + N = black background with white image at normal contrast 12; B + H = black background with white image at maximum contrast 16. Readers analyzed the images to identify morphologic features. Interreader correlation was calculated. Differences between Fleiss-kappa correlation coefficients were examined using bootstrap method. Any setting with significantly higher correlation coefficient was deemed superior for evaluating specific features. Correlation coefficients differed among settings. No single setting was superior for all respective spectral-domain optical coherence tomography parameters (P = 0.3773). Some variables showed no differences among settings. Hard exudates and subretinal fluid were best seen with B + H (κ = 0.46, P = 0.0237 and κ = 0.78, P = 0.002). Microaneurysms were best seen with W + N (κ = 0.56, P = 0.025). Vitreomacular interface, enhanced transmission signal, and epiretinal membrane were best identified using all color/contrast settings together (κ = 0.44, P = 0.042, κ = 0.57, P = 0.01, and κ = 0.62, P ≤ 0.0001). Contrast and background affect the evaluation of retinal structures on spectral-domain optical coherence tomography images. No single setting was superior for all features, though certain changes were best seen with specific settings.