Retinal Features Research Articles

Retinal oculomics and risk of incident aortic aneurysm and aortic adverse events: a population-based cohort study.

The asymptomatic onset and extremely high mortality rate of aortic aneurysm (AA) highlight the urgency of early detection and timely intervention. The alteration of retinal vascular features (RVFs) can reflect the systemic vascular properties, and be widely used as the biomarker for cardiovascular disease risk prediction. Therefore, we aimed to investigate associations of RVFs with AA and its progression. In this prospective population-based cohort study, participants with eligible fundus images and without a history of AA at recruitment were included for analysis. A fully automated Retina-based Microvascular Health Assessment System was used to quantify multidimensional RVFs including the branching angle, caliber, complexity, density, length, and tortuosity. Univariable and multivariable Cox regressions were used to estimate the association of RVFs with the incidence of AA and aortic adverse events (AAE). Furthermore, propensity score matching was performed to mitigate the confounding effects of baseline characteristics. During a median follow-up of 11.0years, 306 incident AA (164 with abdominal AA and 108 with thoracic AA) and 48 incident AAE were documented. In the fully adjusted model, the retinal arterial branching angle (hazard ratio [HR] 0.87, 95% confidence interval [CI] 0.77 to 0.99) and the central tendency and variability of minimum venular caliber were significantly associated with the risk of incident AA (HR 1.13-1.15), while the venular minimum angular asymmetry (0.48, 0.30 to 0.77) was significantly associated with the incidence of AAE. Moreover, specific alterations of RVFs were observed in different AA subtypes (caliber in abdominal AA [HR 1.21]; caliber [HR 1.21-1.28], complexity, length, and tortuosity [HR 0.77-0.82] in thoracic AA). Similar results were obtained after propensity score-matched analysis, confirming the stability of these associations. We identified a significant association of certain RVFs with incident AA and AAE, implying that noninvasive, and convenient fundus photography could be a promising tool to facilitate the early detection of AA and subsequent preventative interventions.

Predicting branch retinal vein occlusion development using multimodal deep learning and pre-onset fundus hemisection images

Branch retinal vein occlusion (BRVO) is a leading cause of visual impairment in working-age individuals, though predicting its occurrence from retinal vascular features alone remains challenging. We developed a deep learning model to predict BRVO based on pre-onset, metadata-matched fundus hemisection images. This retrospective cohort study included patients diagnosed with unilateral BRVO from two Korean tertiary centers (2005–2023), using hemisection fundus images from 27 BRVO-affected eyes paired with 81 unaffected hemisections (27 counter and 54 contralateral) for training. A U-net model segmented retinal optic discs and blood vessels (BVs), dividing them into upper and lower halves labeled for BRVO occurrence. Both unimodal models (using either fundus or BV images) and a BV-enhanced multimodal model were constructed to predict future BRVO. The multimodal model outperformed the unimodal models achieving an area under the receiver operating characteristic curve of 0.76 (95% confidence interval [CI], 0.66–0.83) and accuracy of 68.5% (95% CI 58.9–77.1%), with predictions focusing on arteriovenous crossing regions in the retinal vascular arcade. These findings demonstrate the potential of the BV-enhanced multimodal approach for BRVO prediction and highlight the need for larger, multicenter datasets to improve its clinical utility and predictive accuracy.

Intraoperative Augmented Reality for Vitreoretinal Surgery Using Edge Computing.

Purpose: Augmented reality (AR) may allow vitreoretinal surgeons to leverage microscope-integrated digital imaging systems to analyze and highlight key retinal anatomic features in real time, possibly improving safety and precision during surgery. By employing convolutional neural networks (CNNs) for retina vessel segmentation, a retinal coordinate system can be created that allows pre-operative images of capillary non-perfusion or retinal breaks to be digitally aligned and overlayed upon the surgical field in real time. Such technology may be useful in assuring thorough laser treatment of capillary non-perfusion or in using pre-operative optical coherence tomography (OCT) to guide macular surgery when microscope-integrated OCT (MIOCT) is not available. Methods: This study is a retrospective analysis involving the development and testing of a novel image-registration algorithm for vitreoretinal surgery. Fifteen anonymized cases of pars plana vitrectomy with epiretinal membrane peeling, along with corresponding preoperative fundus photographs and optical coherence tomography (OCT) images, were retrospectively collected from the Mayo Clinic database. We developed a TPU (Tensor-Processing Unit)-accelerated CNN for semantic segmentation of retinal vessels from fundus photographs and subsequent real-time image registration in surgical video streams. An iterative patch-wise cross-correlation (IPCC) algorithm was developed for image registration, with a focus on optimizing processing speeds and maintaining high spatial accuracy. The primary outcomes measured were processing speed in frames per second (FPS) and the spatial accuracy of image registration, quantified by the Dice coefficient between registered and manually aligned images. Results: When deployed on an Edge TPU, the CNN model combined with our image-registration algorithm processed video streams at a rate of 14 FPS, which is superior to processing rates achieved on other standard hardware configurations. The IPCC algorithm efficiently aligned pre-operative and intraoperative images, showing high accuracy in comparison to manual registration. Conclusions: This study demonstrates the feasibility of using TPU-accelerated CNNs for enhanced AR in vitreoretinal surgery.

Retinal OCT-Derived Texture Features as Potential Biomarkers for Early Diagnosis and Progression of Diabetic Retinopathy.

Diabetic retinopathy (DR) is usually diagnosed many years after diabetes onset. Indeed, an early diagnosis of DR remains a notable challenge, and, thus, developing novel approaches for earlier disease detection is of utmost importance. We aim to explore the potential of texture analysis of optical coherence tomography (OCT) retinal images in detecting retinal changes in streptozotocin (STZ)-induced diabetic animals at "silent" disease stages when early retinal molecular and cellular changes that cannot be clinically detectable are already occurring. Volume OCT scans and electroretinograms were acquired before and 1, 2, and 4 weeks after diabetes induction. Automated OCT image segmentation was performed, followed by retinal thickness and texture analysis. Blood-retinal barrier breakdown, glial reactivity, and neuroinflammation were also assessed. Type 1 diabetes induced significant early changes in several texture metrics. At week 4 of diabetes, autocorrelation, correlation, homogeneity, information measure of correlation II (IMCII), inverse difference moment normalized (IDN), inverse difference normalized (INN), and sum average texture metrics decreased in all retinal layers. Similar effects were observed for correlation, homogeneity, IMCII, IDN, and INN at week 2. Moreover, the values of those seven-texture metrics described above decreased throughout the disease progression. In diabetic animals, subtle retinal thinning and impaired retinal function were detected, as well as an increase in the number of Iba1-positive cells (microglia/macrophages) and a subtle decrease in the tight junction protein immunoreactivity, which did not induce any physiologically relevant effect on the blood-retinal barrier. The effects of diabetes on the retina can be spotted through retinal texture analysis in the early stages of the disease. Changes in retinal texture are concomitant with biological retinal changes, thus unlocking the potential of texture analysis for the early diagnosis of DR. However, this requires to be proven in clinical studies.

Noninvasive Anemia Detection and Hemoglobin Estimation from Retinal Images Using Deep Learning: A Scalable Solution for Resource-Limited Settings.

The purpose of this study was to develop and validate a deep-learning model for noninvasive anemia detection, hemoglobin (Hb) level estimation, and identification of anemia-related retinal features using fundus images. The dataset included 2265 participants aged 40years and above from a population-based study in South India. The dataset included ocular and systemic clinical parameters, dilated retinal fundus images, and hematological data such as complete blood counts and Hb concentration levels. Eighty percent of the dataset was used for algorithm development and 20% for validation. A deep-convolutional neural network, utilizing VGG16, ResNet50, and InceptionV3 architectures, was trained to predict anemia and estimate Hb levels. Sensitivity, specificity, and accuracy were calculated, and receiver operating characteristic (ROC) curves were generated for comparison with clinical anemia data. GradCAM saliency maps highlighted regions linked to anemia and image processing techniques to quantify anemia-related features. For predicting anemia, the InceptionV3 model demonstrated the best performance, achieving 98% accuracy, 99% sensitivity, 97% specificity, and an area under the curve (AUC) of 0.98 (95% confidence interval [CI] = 0.97-0.99). For estimating Hb levels, the mean absolute error for the InceptionV3 model was 0.58 g/dL (95% CI = 0.57-0.59 g/dL). The model focused on the area around the optic disc and the neighboring retinal vessels, revealing that anemic subjects exhibited significantly increased vessel tortuosity and reduced vessel density (P < 0.001), with variable effects on vessel thickness. The InceptionV3 model accurately predicted anemia and Hb levels, highlighting the potential of deep learning and vessel analysis for noninvasive anemia detection. The proposed method offers the possibility to quantitatively predict hematological parameters in a noninvasive manner.

Automated analysis of retinal vascular features, age, sex and disc size in a large cohort of primary open angle glaucoma patients

Aims/Purpose: Deriving vascular features of retinal images is a proposed noninvasive method to assess vascular health. Although several studies have linked cardiovascular risk to retinal image features, they often rely on limited datasets or have used deep learning approaches with limited explainability. This research introduces an end‐to‐end method for analyzing the retinal vasculature in large retinal image datasets, applied for primary open angle glaucoma (POAG).Methods: 115,237 retinal images of the UZ Leuven Glaucoma Clinic were extracted. 4858 unique images remain of POAG patients (n = 3010) and controls (n = 1848) after image quality assessment, optic disk detection, region of interest definition, automated segmentation of arterioles and venules (LUNet algorithm), and parametrization of the vascular biomarkers (PVBM toolbox). Analysis of covariance and linear mixed models were used to adjust for age, sex and disc size.Results: Both arteriolar and venular diameter, area, length, tortuosity, branching angle, endpoints, intersection points and both mono‐ as multifractal dimension levels are independently lower in POAG patients and older patients (all p &lt; 0.001), after adjustment for sex, disc size and multiple testing. The multivariate linear mixed models additionally show that the vascular features are significantly influenced by sex and disc size, which necessitates correct adjustment.Conclusions: This is the first time a fully automated end‐to‐end pipeline is published for the analysis of retinal vascular geometry in a large cohort (n = 4858). Given the independent similarities in retinal vascular geometry changes between older age and POAG prevalence the theory of pronounced vascular ageing in POAG patients is proposed. This novel approach enables quantitative analysis of eye vasculature and supports the study of its correlation with specific diseases, facilitating reproducible analysis of large datasets and providing explainability to deep learning approaches.

Detection of Retinal and Choriocapillaris Microvascular Changes in Retinal Vein Occlusion and Fellow Eyes by Optical Coherence Tomography Angiography: A Systematic Review and Meta-Analysis.

This study aims to summarize the retinal and choroidal microvascular features detected by optical coherence tomography angiography (OCTA) in the affected and fellow eyes of patients with retinal vein occlusion (RVO). A comprehensive search of the PubMed, Embase, and Ovid databases was conducted to identify studies comparing OCTA metrics among RVO, RVO-fellow, and control eyes. Outcomes of interest included parameters related to foveal avascular zone (FAZ) and fovea- and optic nerve head (ONH)-centered perfusion measurements of superficial capillary plexus (SCP), deep capillary plexus (DCP), and choriocapillaris layer. Pooled results were presented as mean differences or standardized mean differences with 95% confidence intervals. Fifty-three studies, comprising 2119 RVO, 1393 fellow, and 1178 control eyes, were included in the quantitative meta-analysis. RVO eyes exhibited larger FAZ areas, increased FAZ acircularity, and reduced macular retinal and choriocapillaris perfusion compared to RVO-fellow and control eyes (P < 0.05). RVO eyes also demonstrated significantly lower perfusion density (PD) in the inside-disk and peripapillary regions of the radial peripapillary capillary layer (RPC), as well as lower retinal and choriocapillaris PD in the 4.5 × 4.5mm2 field of view (FOV) of ONH-centered scans (P < 0.05). RVO-fellow eyes showed decreased SCP-PD and DCP-PD in theparafoveal region and the 3 × 3mm2 FOV, reduced inside-disk and 4.5 × 4.5mm2 FOV RPC-PD (P < 0.05), and a diminished choriocapillaris flow area in the 3 × 3mm2 FOV (P < 0.05). Both RVO-affected and RVO-fellow eyes exhibited retinal and choriocapillaris microvascular impairment around the fovea and ONH. OCTA represents a promising tool for comprehensively assessing vascular alterations in RVO and providing evidence of fellow eye involvement.

A Customized CNN Architecture with CLAHE for Multi-Stage Diabetic Retinopathy Classification

This paper presents a customized Convolutional Neural Network (CNN) architecture for multi-stage detection of Diabetic Retinopathy (DR), a leading cause of vision impairment and blindness. The proposed model incorporates advanced image enhancement techniques, particularly Contrast Limited Adaptive Histogram Equalization (CLAHE), to improve the visibility of critical retinal features associated with DR. By integrating CLAHE with a finely tuned CNN, the proposed approach significantly enhances accuracy and robustness, allowing for more precise detection across various stages of DR. The proposed model was evaluated against several state-of-the-art techniques, with the customized CNN alone achieving an overall accuracy of 97.69%. The addition of CLAHE further boosts the performance, achieving an accuracy of 99.69%, underscoring the effectiveness of combining CLAHE with CNN for automated DR detection. This approach provides an efficient, scalable, and highly accurate solution for early and multistage DR detection, which is crucial for timely intervention and prevention of vision loss.

Relationship between autism spectrum disorder and peripapillary intraretinal layer thickness: a pediatric retrospective cross-sectional study.

Autism spectrum disorder (ASD) often presents with atypical visual processing, prompting investigation into its connection with retinal features. This study aimed to (I) compare intraretinal layer thickness in the peripapillary region between ASD and neurotypical (NT) groups, (II) assess associations between intraretinal layer thickness and clinical parameters (social functioning and cognitive levels) in ASD subjects, and (III) evaluate the potential of intraretinal layer thickness as a biomarker for ASD. Participants were recruited through convenience sampling from the Children's Mental Health Research Center at The Affiliated Brain Hospital of Nanjing Medical University and the Department of Ophthalmology at The First Affiliated Hospital of Nanjing Medical University, Nanjing, China, between December 2019 and August 2023. Intraretinal layer thickness in peripapillary region was quantified using optic coherence tomography images with automated layer segmentation performed by OCTExplorer software on 47 individuals with ASD (aged 7-13 years) and age- and sex-matched NT controls. Inter-group comparisons were conducted using unpaired t-tests, Welch's t-tests, or Mann-Whitney U tests as appropriate. Correlations with social functioning (measured by Social Responsiveness Scale scores) and cognitive levels [measured by total intelligence quotient (IQ) scores] were examined using the Spearman correlation coefficient. Stepwise regression analysis was conducted to assess predictive power. Significant inter-group differences (P<0.05) were observed in ganglion cell layer and inner nuclear layer (INL) thickness across global and specific quadrant regions. Participants had a mean age of 9.57±1.83 years in the ASD group and 9.89±1.70 years in the age-matched NT group. While no correlation was found between retinal sublayer thickness and social functioning on ASD subjects (all P>0.05), there was a notable correlation between INL thickness in the infero-nasal quadrant and cognitive level (r=0.381, P=0.014). Stepwise regression analysis identified global INL thickness as a significant predictor of total IQ scores (β=3.986, P=0.034), with an R2 of 0.110 and a root mean square error of 21.900. This study highlights significant differences in retinal features between ASD and NT groups, with implications for understanding ASD pathogenesis and complexity. The findings suggest that easily observable retinal features hold promise as biomarkers for ASD, warranting further investigation.

Homonymous hemi-macular atrophy in multiple sclerosis

Background: Retrograde trans-synaptic degeneration (TSD) following retro-chiasmal pathology, typically retro-geniculate in multiple sclerosis (MS), may manifest as homonymous hemi-macular atrophy (HHMA) of the ganglion cell/inner plexiform layer (GCIPL). Objective: To determine the frequency, association with clinical outcomes, and retinal and radiological features of HHMA in people with MS (PwMS). Methods: In this cross-sectional study, healthy controls (HC) and PwMS underwent retinal optical coherence tomography scanning. For quantitative identification of HHMA, a normalized asymmetry ratio was used, and its normative cutoffs were established from the HC. HHMA PwMS were propensity score matched 1:2 to non-HHMA PwMS. Mixed-effects linear regression models were used in analyses. Results: Based on normative data from 238 HC (466 eyes), 79 out of 942 PwMS exhibited HHMA (8.4%; 143 eyes). Compared to non-HHMA eyes from matched PwMS (158 PwMS; 308 eyes), HHMA eyes had lower average GCIPL (diff: −5.7 μm (95% CI −7.6 to −3.8); p < 0.001) but also inner nuclear layer (diff: −0.9 μm (95% CI −1.6 to −0.1); p = 0.02), and outer nuclear layer (diff: −1.9 μm (95% CI −3.4 to −0.4); p = 0.02) thicknesses; in further analyses, these differences were exclusive to the homonymous side of HHMA. HHMA participants also exhibited higher expanded disability status scale scores (diff: 0.5 (95% CI 0.1 to 0.9); p = 0.02), worse 100% and 2.5% visual acuity scores (diff: −3.2 (95% CI −4.1 to −1.0); p = 0.002, −5.4 (95% CI −7.5 to −3.5); p < 0.001), and higher frequency of microcystoid macular changes (10.1% vs. 3.2%; p = 0.03) compared to non-HHMA participants. Conclusions: HHMA, possibly as a marker of TSD, may signify higher disability in MS.

Extracting full information from OCT scans-signs of early age-related macular degeneration within inner retinal layers by local neighbourhood statistics. Part II: Results.

Associations between the occurrence of early age related macular degeneration (AMD) and alterations in retinal layer thicknesses have been reported, based on classical processing of optical coherence tomography (OCT) data by noise removal and subsequent image segmentation. However, speckle noise within OCT data itself bears a substantial part of the total information. For this reason, we designed an omics-type approach for full exploitation of OCT data, which was able to identify signs of early AMD throughout the retina as a whole. A nested case-control study was designed with 200 early AMD cases and 200 healthy controls. For each participant, within a randomly selected OCT scan and a randomly selected column therein, manual grading was performed for 26 retinal feature positions. At every position, a total of 3792 descriptors were computed, based on nonlinear transformations of OCT data, first-order neighbourhood statistics and Haralick features. Equivalence and differences between cases and controls were tested for each descriptor at every graded position. Results of multiple testing were expressed in terms of false and true discovery rates controlled by the Benjamini-Yekutieli procedure. In terms of the amount and disparity of true discoveries, overall non-equivalence was found for early AMD and healthy groups. Strong difference signals were observed at the internal limiting membrane and two central retinal positions, particularly for descriptors emphasising speckle noise. Between the retinae of healthy controls and early AMD patients, significant differences were observed at the level of local neighbourhood statistics within OCT data. Thus, independent evidence was obtained for AMD affecting not only the outer retinal layers but the retina as a whole, even in the early stages of the disease. Within OCT data, both cartoon and speckle bear essential parts of the total information. We pursued a constructive, completely documented, traceable and repeatable approach without invoking artificial intelligence methods.

Extracting full information from OCT scans-signs of early age-related macular degeneration within inner retinal layers by local neighbourhood statistics. Part I: Methodology.

Associations between the occurrence of early age-related macular degeneration (AMD) and alterations in retinal layer thicknesses have been reported based on classical processing of optical coherence tomography (OCT) data by noise removal and subsequent image segmentation. However, speckle noise within OCT data itself bears a substantial part of the total information. For this reason, an omics-type approach was designed for full exploitation of OCT data, which was able to identify signs of early AMD throughout the retina as a whole. A nested case-control study was designed with 200 early AMD cases and 200 healthy controls. For every participant, within a randomly selected OCT scan and a randomly selected column therein, manual grading was performed for 26 retinal feature positions. At each position, a total of 3792 descriptors were computed, based on nonlinear transformations of OCT data, first-order neighbourhood statistics and Haralick features. Equivalence and differences between cases and controls were tested for every descriptor at each graded position. Results of multiple testing were expressed in terms of false and true discovery rates controlled by the Benjamini-Yekutieli procedure. In terms of the amount and disparity of true discoveries, overall non-equivalence of early AMD and healthy groups was found. Strong difference signals were observed at the internal limiting membrane and two central retinal positions, particularly for descriptors emphasising speckle noise. Between retinae of healthy controls and early AMD patients, significant differences were observed at the level of local neighbourhood statistics within the OCT data. Thus, independent evidence was obtained for AMD affecting not only the outer retinal layers but also the retina as a whole, even in the early stages of the disease. Within OCT data, both cartoons and speckle bear essential parts of total information. A constructive, completely documented, traceable and repeatable approach was pursued without invoking artificial intelligence methods.

Retinal structure guidance-and-adaption network for early Parkinson’s disease recognition based on OCT images

Parkinson’s disease (PD) is a leading neurodegenerative disease globally. Precise and objective PD diagnosis is significant for early intervention and treatment. Recent studies have shown significant correlations between retinal structure information and PD based on optical coherence tomography (OCT) images, providing another potential means for early PD recognition. However, how to exploit the retinal structure information (e.g., thickness and mean intensity) from different retinal layers to improve PD recognition performance has not been studied before. Motivated by the above observations, we first propose a structural prior knowledge extraction (SPKE) module to obtain the retinal structure feature maps; then, we develop a structure-guided-and-adaption attention (SGDA) module to fully leverage the potential of different retinal layers based on the extracted retinal structure feature maps. By embedding SPKE and SGDA modules at the low stage of deep neural networks (DNNs), a retinal structure-guided-and-adaption network (RSGA-Net) is constructed for early PD recognition based on OCT images. The extensive experiments on a clinical OCT-PD dataset demonstrate the superiority of RSGA-Net over state-of-the-art methods. Additionally, we provide a visual analysis to explain how retinal structure information affects the decision-making process of DNNs.

Choroidal and retinal alteration after long-term use of tadalafil: a prospective non-randomized clinical trial.

The purpose of this study is to investigate choroidal and retinal vascular features in patients taking PDE5is by measuring dynamic vascular alterations and neurostructural features of the retina before and after oral tadalafil administration. The current clinical research involved 22 patients treated with tadalafil 20 mg on alternate days (OAD) after nerve-sparing robotic radical prostatectomy (NS-RARP) for prostate cancer. Patients underwent SD-OCT to assess ganglion cell complex (GCC), retinal nerve fiber layer (RNFL), and subfoveal choroidal thickness (SFCT), as well as OCTA to assess superficial capillary plexus (SCP), deep capillary plexus (DCP), choriocapillaris (CC), foveal avascular zone (FAZ), and radial peripapillary capillary thickness (RPC). All patients were evaluated at baseline (t0), and 3 (t1) and 6 (t2) months after the use of oral tadalafil. A statistically significant reduction in DCP and CC vessel density was found at t2 compared to baseline. According to the SFCT parameter, a statistically significant increase was observed from t0 to t1, and from t1 to t2. GCC parameter increased at t2 compared to baseline in a statistically significant way. No statistically significant differences were recorded between t0, t1 and t2 for the SCP, RPC, FAZ area, RNFL parameter. Retinal and optic disc toxicity may be detected using modifications of capillary vessel density. Further studies are needed to detect the possible progression or regression of ocular or systemic vascular complications in long-term follow-up.

Developing a 10-Layer Retinal Segmentation for MacTel Using Semi-Supervised Learning.

Automated segmentation software in optical coherence tomography (OCT) devices is usually developed for and primarily tested on common diseases. Therefore segmentation accuracy of automated software can be limited in eyes with rare pathologies. We sought to develop a semisupervised deep learning segmentation model that segments 10 retinal layers and four retinal features in eyes with Macular Telangiectasia Type II (MacTel) using a small labeled dataset by leveraging unlabeled images. We compared our model against popular supervised and semisupervised models, as well as conducted ablation studies on the model itself. Our model significantly outperformed all other models in terms of intersection over union on the 10 retinal layers and two retinal features in the test dataset. For the remaining two features, the pre-retinal space above the internal limiting membrane and the background below the retinal pigment epithelium, all of the models performed similarly. Furthermore, we showed that using more unlabeled images improved the performance of our semisupervised model. Our model improves segmentation performance over supervised models by leveraging unlabeled data. This approach has the potential to improve segmentation performance for other diseases, where labeled data is limited but unlabeled data abundant. Improving automated segmentation of MacTel pathology on OCT imaging by leveraging unlabeled data may enable more accurate assessment of disease progression, and this approach may be useful for improving feature identification and location on OCT in other rare diseases as well.

Validation of Inter-Reader Agreement/Consistency for Quantification of Ellipsoid Zone Integrity and Sub-RPE Compartmental Features Across Retinal Diseases.

An unmet need exists when clinically assessing retinal and layer-based features of retinal diseases. Therefore, quantification of retinal-layer-thicknesses/fluid volumes using deep-learning-augmented platforms to reproduce human-obtained clinical measurements is needed. In this analysis, 210 spectral-domain optical coherence tomography (SD-OCT) scans (30 without pathology, 60 dry age-related macular degeneration [AMD], 60 wet AMD, and 60 diabetic macular edema [total 23,625 B-scans]) were included. A fully automated segmentation platform segmented four retinal layers for compartmental assessment (internal limiting membrane, ellipsoid zone [EZ], retinal pigment epithelium [RPE], and Bruch's membrane). Two certified OCT readers independently completed manual segmentation and B-scan level validation of automated segmentation, with segmentation correction when needed (semi-automated). Certified reader metrics were compared to gold standard metrics using intraclass correlation coefficients (ICCs) to assess overall agreement. Across different diseases, several metrics generated from automated segmentations approached or matched human readers performance. Absolute ICCs for retinal mean thickness measurements showed excellent agreement (range 0.980-0.999) across four cohorts. EZ-RPE thickness values and sub-RPE compartment ICCs demonstrated excellent agreement (ranges of 0.953-0.987 and 0.944-0.997, respectively) for full dataset, dry-AMD, and wet-AMD cohorts. Analyses demonstrated high reliability and consistency of segmentation of outer retinal compartmental features using a completely human/manual approach or a semi-automated approach to segmentation. These results support the critical role that measuring features, such as photoreceptor preservation through EZ integrity, in future clinical trials may optimize clinical care.

Longitudinal Assessment of Retinopathy of Prematurity (LONGROP) Study: Impacts of Viewing Time and Ability to Compare on Detection of Change

This study compared two imaging grading techniques to assess the utility of longitudinal image-based analysis in retinopathy of prematurity (ROP) screening: (1) time-limited without image comparison (a proxy for bedside indirect ophthalmoscopy, termed sBIO) and time-unlimited with image comparison (for telemedicine grading, termed TELE) screening. We tested two hypotheses: (1) H1: TELE was superior to sBIO for the detection of change (Tempo)-same, better, or worse and (2) H2: granular data of change (e.g., at the image and feature level) is integrated by graders to achieve the Tempo assessment. Prospective reliability analysis. Gold standard reference (GS) was a published curated ROP image database consisting of both Tempo and granular level changes (image and components) from 40 patients in 2 sets. Graders were divided into 2 cohorts. There were two screening techniques: (1) sBIO with time limited review of 10 minutes/patient, access to prior notes and drawings and (2) TELE with unlimited review time, access to prior weeks' images, notes and schematics. Graders switched techniques and sets after 6 weeks. H1 outcome was comparison of graders' weekly Tempo scores to GS-Gestalt and for H2 was Tempo score compared to GS-View and GS-Component. H1 demonstrated no difference-accuracy of sBIO and TELE compared to GS was 51.7% and 51.9% respectively (P = .95). Highest agreement occurred when all exams exhibited no change (91.5% sBIO vs 93.5% TELE, P = .46) and worst agreement was when exams always demonstrated worsening (46.5% sBIO vs 47.1% TELE, P = .93). Both sets of graders did worse in weeks 7-12, irrespective of technique. H2 demonstrated that Tempo assessment did not correlate with granular data changes in the GS for View level and Component level assessments-overall agreement dropped to 31.4% for Tempo vs GS-VIEW (31.2% for sBIO, 31.5% for TELE) and 4.6% for Tempo vs GS-COMPONENT (4.9% for sBIO, 4.3% for TELE). Detection of ROP Tempo was independent of screening technique by expert pediatric retina graders. Both groups did significantly better in the first half of the study, indicative of a fatigue factor. This is the first study in ROP history to demonstrate that graders integrate image and retinal features in various ways that can be in contradiction of their assessment of overall disease progression.

GWAS-by-subtraction reveals an IOP-independent component of primary open angle glaucoma

The etiology of primary open angle glaucoma is constituted by both intraocular pressure-dependent and intraocular pressure-independent mechanisms. However, GWASs of traits affecting primary open angle glaucoma through mechanisms independent of intraocular pressure remains limited. Here, we address this gap by subtracting the genetic effects of a GWAS for intraocular pressure from a GWAS for primary open angle glaucoma to reveal the genetic contribution to primary open angle glaucoma via intraocular pressure-independent mechanisms. Seventeen independent genome-wide significant SNPs were associated with the intraocular pressure-independent component of primary open angle glaucoma. Of these, 7 are located outside known normal tension glaucoma loci, 11 are located outside known intraocular pressure loci, and 2 are novel primary open angle glaucoma loci. The intraocular pressure-independent genetic component of primary open angle glaucoma is associated with glaucoma endophenotypes, while the intraocular pressure-dependent component is associated with blood pressure and vascular permeability. A genetic risk score for the intraocular pressure-independent component of primary open angle glaucoma is associated with 26 different retinal micro-vascular features, which contrasts with the genetic risk score for the intraocular pressure-dependent component. Increased understanding of these intraocular pressure-dependent and intraocular pressure-independent components provides insights into the pathogenesis of glaucoma.

The Optical Nature of Myopic Changes in Retinal Vessel Caliber

PurposeDimensional measures of retinal features are subject to the optical influence of ocular magnification. We examined the impact of ocular magnification on the association between axial length (AL) and measurements of retinal vessel calibre in fundus photographs. DesignCross-sectional study. ParticipantsEighty-two normal right eyes from healthy participants aged 16-31 years. MethodsCentral retinal arteriolar and venular equivalents (CRAE and CRVE) were derived from colour fundus photographs using semi-automated software. Ordinary least squares linear regression was used to assess the influence of AL (independent variable) on CRAE and CRVE, controlling for age, sex and ethnicity, both before and after magnification correction using different formulae. These formulae estimate magnification based on different ocular parameters: AL only (Bennnett’s formula), refractive error only (Bengtsson’s formula) and refractive error combined with keratometry (Littmann’s formula). Previous research has primarily relied on Bengtsson’s formula, which is less accurate than Bennett’s formula. We also examined the impact of treating the non-telecentric fundus camera used in this study as telecentric when applying these magnification correction formulae. Main Outcome MeasuresCentral retinal arteriolar and venular equivalents (in pixels). ResultsBefore magnification correction, increasing AL was associated with decreasing CRAE (β: -0.49, 95% CI: -0.89 to -0.09, p=0.02) and CRVE (β: -0.91, 95% CI: -1.62 to -0.20, p=0.01). After magnification correction, this observation was no longer evident, regardless of the correction formula applied. When inappropriately assuming the fundus camera to be telecentric, we observed a bias towards increasing magnification-corrected CRAE and CRVE with increasing AL (β coefficients were positive or became more positive), reaching statistical significance (p<0.05) for CRAE corrected using Bennett’s or Littmann’s formula, and for CRVE corrected using Bennett’s formula. ConclusionsFailing to correct for ocular magnification results in apparent narrowing of vessels in longer eyes, while inappropriate assumptions about telecentricity during magnification correction introduce an optical artifact that causes apparent widening of vessels. These findings suggest that myopic changes in retinal vessel calibre are optical (not biological) in nature. Proper correction of this effect to accurately derive dimensional measures is a crucial—yet often overlooked—methodological consideration in “oculomics” research investigating retinal biomarkers of systemic conditions.

Adaptive optics scanning laser ophthalmoscopy in a heterogenous cohort with Stargardt disease

Image based cell-specific biomarkers will play an important role in monitoring treatment outcomes of novel therapies in patients with Stargardt (STGD1) disease and may provide information on the exact mechanism of retinal degeneration. This study reports retinal image features from conventional clinical imaging and from corresponding high-resolution imaging with a confocal adaptive optics scanning laser ophthalmoscope (AOSLO) in a heterogenous cohort of patients with Stargardt (STGD1) disease. This is a prospective observational study in which 16 participants with clinically and molecularly confirmed STGD1, and 7 healthy controls underwent clinical assessment and confocal AOSLO imaging. Clinical assessment included short-wavelength and near-infrared fundus autofluorescence, spectral-domain optical coherence tomography, and macular microperimetry. AOSLO images were acquired over a range of retinal eccentricities (0°–20°) and mapped to areas of interest from the clinical images. A regular photoreceptor mosaic was identified in areas of normal or near normal retinal structure on clinical images. Where clinical imaging indicated areas of retinal degeneration, the photoreceptor mosaic was disorganised and lacked unambiguous cones. Discrete hyper-reflective foci were identified in 9 participants with STGD1 within areas of retinal degeneration. A continuous RPE cell mosaic at the fovea was identified in one participant with an optical gap phenotype. The clinical heterogeneity observed in STGD1 is reflected in the findings on confocal AOSLO imaging.