The Association of Urinary Sodium Excretion with Glaucoma and Related Traits in a Large United Kingdom Population.

OBJECTIVE: Excessive dietary sodium intake has known adverse effects on intravascular fluid volume and systemic blood pressure, which may influence intraocular pressure (IOP) and glaucoma risk. This study aimed to assess the association of urinary sodium excretion, a biomarker of dietary intake, with glaucoma and related traits, and to determine whether this relationship is modified by genetic susceptibility to disease. DESIGN: Cross-sectional observational and gene-environment interaction analyses in the population-based UK Biobank study. PARTICIPANTS: Up to 103 634 individuals (mean age 57 years, 51% women) with complete urinary, ocular, and covariable data. METHODS: Urine sodium:creatinine ratio (UNa:Cr; mmol:mmol) was calculated from a midstream urine sample. Ocular parameters were measured as part of a comprehensive eye examination and glaucoma case ascertainment was through a combination of self-report and linked national hospital records. Genetic susceptibility to glaucoma was calculated based on a glaucoma polygenic risk score (PRS) comprising 2 673 common genetic variants. Multivariable linear and logistic regression, adjusted for key sociodemographic, medical, anthropometric, and lifestyle factors, were used to model associations and gene-environment interactions. MAIN OUTCOME MEASURES: Corneal-compensated IOP, optical coherence tomography derived macular retinal nerve fiber layer (mRNFL) and ganglion cell-inner plexiform layer (GCIPL) thickness, and prevalent glaucoma. RESULTS: In maximally adjusted regression models, a one standard deviation increase in UNa:Cr was associated with higher IOP (0.14mmHg; 95% CI, 0.12 to 0.17; P<0.001) and greater prevalence of glaucoma (OR, 1.11; 95% CI, 1.07 to 1.14; P<0.001), but not mRNFL or GCIPL thickness. Compared to those with UNa:Cr in the lowest quintile, those in the highest quintile had significantly higher IOP (0.45mmHg; 95% CI, 0.36 to 0.53, P<0.001) and prevalence of glaucoma (OR, 1.30; 95% CI, 1.17 to 1.45; P<0.001). Stronger associations with glaucoma (P interaction=0.001) were noted in participants with a higher glaucoma PRS. CONCLUSIONS: Urinary sodium excretion, a biomarker of dietary intake, may represent an important modifiable risk factor for glaucoma, especially in individuals at high underlying genetic risk. These findings warrant further investigation as they may have important clinical and public health implications.

Periodontitis and Outer Retinal Thickness: a Cross-Sectional Analysis of the United Kingdom Biobank Cohort.

Purpose: Periodontitis, a ubiquitous severe gum disease affecting the teeth and surrounding alveolar bone, can heighten systemic inflammation. We investigated the association between very severe periodontitis and early biomarkers of age-related macular degeneration (AMD), in individuals with no eye disease. Design: Cross-sectional analysis of the prospective community-based cohort United Kingdom (UK) Biobank. Participants: Sixty-seven thousand three hundred eleven UK residents aged 40 to 70 years recruited between 2006 and 2010 underwent retinal imaging. Methods: Macular-centered OCT images acquired at the baseline visit were segmented for retinal sublayer thicknesses. Very severe periodontitis was ascertained through a touchscreen questionnaire. Linear mixed effects regression modeled the association between very severe periodontitis and retinal sublayer thicknesses, adjusting for age, sex, ethnicity, socioeconomic status, alcohol consumption, smoking status, diabetes mellitus, hypertension, refractive error, and previous cataract surgery. Main Outcome Measures: Photoreceptor layer (PRL) and retinal pigment epithelium-Bruch's membrane (RPE-BM) thicknesses. Results: Among 36 897 participants included in the analysis, 1571 (4.3%) reported very severe periodontitis. Affected individuals were older, lived in areas of greater socioeconomic deprivation, and were more likely to be hypertensive, diabetic, and current smokers (all P < 0.001). On average, those with very severe periodontitis were hyperopic (0.05 ± 2.27 diopters) while those unaffected were myopic (-0.29 ± 2.40 diopters, P < 0.001). Following adjusted analysis, very severe periodontitis was associated with thinner PRL (-0.55 µm, 95% confidence interval [CI], -0.97 to -0.12; P = 0.022) but there was no difference in RPE-BM thickness (0.00 µm, 95% CI, -0.12 to 0.13; P = 0.97). The association between PRL thickness and very severe periodontitis was modified by age (P < 0.001). Stratifying individuals by age, thinner PRL was seen among those aged 60 to 69 years with disease (-1.19 µm, 95% CI, -1.85 to -0.53; P < 0.001) but not among those aged < 60 years. Conclusions: Among those with no known eye disease, very severe periodontitis is statistically associated with a thinner PRL, consistent with incipient AMD. Optimizing oral hygiene may hold additional relevance for people at risk of degenerative retinal disease. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Oculomics: A Crusade Against the Four Horsemen of Chronic Disease.

Chronic, non-communicable diseases present a major barrier to living a long and healthy life. In many cases, early diagnosis can facilitate prevention, monitoring, and treatment efforts, improving patient outcomes. There is therefore a critical need to make screening techniques as accessible, unintimidating, and cost-effective as possible. The association between ocular biomarkers and systemic health and disease (oculomics) presents an attractive opportunity for detection of systemic diseases, as ophthalmic techniques are often relatively low-cost, fast, and non-invasive. In this review, we highlight the key associations between structural biomarkers in the eye and the four globally leading causes of morbidity and mortality: cardiovascular disease, cancer, neurodegenerative disease, and metabolic disease. We observe that neurodegenerative disease is a particularly promising target for oculomics, with biomarkers detected in multiple ocular structures. Cardiovascular disease biomarkers are present in the choroid, retinal vasculature, and retinal nerve fiber layer, and metabolic disease biomarkers are present in the eyelid, tear fluid, lens, and retinal vasculature. In contrast, only the tear fluid emerged as a promising ocular target for the detection of cancer. The retina is a rich source of oculomics data, the analysis of which has been enhanced by artificial intelligence-based tools. Although not all biomarkers are disease-specific, limiting their current diagnostic utility, future oculomics research will likely benefit from combining data from various structures to improve specificity, as well as active design, development, and optimization of instruments that target specific disease signatures, thus facilitating differential diagnoses.

Long-term diseases can stop people living long and healthy lives. In many cases, early diagnosis can help to prevent, monitor, and treat disease, which can improve patients' health. In order to diagnose disease, we need tools that are easy for patients to access, painless, and low-cost. The eye may provide the solution. In this review, we discuss the link between changes in the eye and four types of long-term disease that, together, kill most of the population: (1) Cardiovascular disease (affecting the heart and/or blood). (2) Cancer (abnormal growth of cells). (3) Neurodegenerative disease (affecting the brain and/or nervous system). (4) Metabolic disease (problems storing, accessing, and using the body's fuel). We show that neurodegenerative disease leaves tell-tale signs in lots of different parts of the eye. Signs of cardiovascular and metabolic disease biomarkers are mostly found in the back of the eye, and signs of cancer can be found in the tear fluid. Although signs of disease can be seen in the eye, not all of them will tell us what the disease is. We believe that future research will help us to understand more about long-term disease and how to detect it if we combine information from different structures within the eye and develop new tools to target these specific structures.

Relationship between Oral Health and Glaucoma Traits in the United Kingdom.

PRECIS: In this cross-sectional analysis of UK Biobank participants, we find no adverse association between self-reported oral health conditions and either glaucoma or elevated intraocular pressures. PURPOSE: Poor oral health may cause inflammation that accelerates the progression of neurodegenerative diseases. We investigated the relationship between oral health and glaucoma. PATIENTS: United Kingdom (UK) Biobank participants. METHODS: This is a cross-sectional analysis of participants categorized by self-reported oral health status. Multivariable linear and logistic regression models were employed. Primary analysis examined the association with glaucoma prevalence. Secondary analyses examined associations with IOP, macular retinal nerve fiber layer (mRNFL), and ganglion cell inner plexiform layer (mGCIPL) thicknesses, and interaction terms with multi-trait glaucoma polygenic risk scores (MTAG PRS) or intraocular pressure (IOP) PRS. RESULTS: 170,815 participants (34.3%) reported current oral health problems, including painful or bleeding gums, toothache, loose teeth, and/or denture wear. 33,059, 33,004, 14,652, and 14,613 participants were available for analysis of glaucoma, IOP, mRNFL, and mGCIPL, respectively. No association between oral health and glaucoma was identified (odds ratio (OR): 1.04, 95% confidence interval (CI): 0.95, 1.14). IOPs were slightly lower among those with oral disease (-0.08 mmHg, 95% CI: -0.15, -0.009); specifically, among those with loose teeth (P=0.03) and denture-wearers (P<0.0001). mRNFL measurements were lower among those with oral health conditions (-0.14 microns, 95% CI: -0.27, -0.0009), but mGCIPL measurements (P=0.96) were not significantly different. A PRS for IOP or glaucoma did not modify relations between oral health and IOP nor glaucoma (P-for-interactions0.17). CONCLUSIONS: Self-reported oral health was not associated with elevated IOP nor increased risk of glaucoma. Future studies should confirm the null association between clinically diagnosed oral health conditions and glaucoma.

CF-Loss: Clinically-relevant feature optimised loss function for retinal multi-class vessel segmentation and vascular feature measurement.

Characterising clinically-relevant vascular features, such as vessel density and fractal dimension, can benefit biomarker discovery and disease diagnosis for both ophthalmic and systemic diseases. In this work, we explicitly encode vascular features into an end-to-end loss function for multi-class vessel segmentation, categorising pixels into artery, vein, uncertain pixels, and background. This clinically-relevant feature optimised loss function (CF-Loss) regulates networks to segment accurate multi-class vessel maps that produce precise vascular features. Our experiments first verify that CF-Loss significantly improves both multi-class vessel segmentation and vascular feature estimation, with two standard segmentation networks, on three publicly available datasets. We reveal that pixel-based segmentation performance is not always positively correlated with accuracy of vascular features, thus highlighting the importance of optimising vascular features directly via CF-Loss. Finally, we show that improved vascular features from CF-Loss, as biomarkers, can yield quantitative improvements in the prediction of ischaemic stroke, a real-world clinical downstream task. The code is available at https://github.com/rmaphoh/feature-loss.

Cross-modality Labeling Enables Noninvasive Capillary Quantification as a Sensitive Biomarker for Assessing Cardiovascular Risk.

Purpose: We aim to use fundus fluorescein angiography (FFA) to label the capillaries on color fundus (CF) photographs and train a deep learning model to quantify retinal capillaries noninvasively from CF and apply it to cardiovascular disease (CVD) risk assessment. Design: Cross-sectional and longitudinal study. Participants: A total of 90732 pairs of CF-FFA images from 3893 participants for segmentation model development, and 49229 participants in the UK Biobank for association analysis. Methods: We matched the vessels extracted from FFA and CF, and used vessels from FFA as labels to train a deep learning model (RMHAS-FA) to segment retinal capillaries using CF. We tested the model's accuracy on a manually labeled internal test set (FundusCapi). For external validation, we tested the segmentation model on 7 vessel segmentation datasets, and investigated the clinical value of the segmented vessels in predicting CVD events in the UK Biobank. Main Outcome Measures: Area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity for segmentation. Hazard ratio (HR; 95% confidence interval [CI]) for Cox regression analysis. Results: On the FundusCapi dataset, the segmentation performance was AUC = 0.95, accuracy = 0.94, sensitivity = 0.90, and specificity = 0.93. Smaller vessel skeleton density had a stronger correlation with CVD risk factors and incidence (P < 0.01). Reduced density of small vessel skeletons was strongly associated with an increased risk of CVD incidence and mortality for women (HR [95% CI] = 0.91 [0.84-0.98] and 0.68 [0.54-0.86], respectively). Conclusions: Using paired CF-FFA images, we automated the laborious manual labeling process and enabled noninvasive capillary quantification from CF, supporting its potential as a sensitive screening method for identifying individuals at high risk of future CVD events. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Smoking, Corneal Biomechanics, and Glaucoma: Results From Two Large Population-Based Cohorts.

Purpose: Smoking may influence measured IOP through an effect on corneal biomechanics, but it is unclear whether this factor translates into an increased risk for glaucoma. This study aimed to examine the association of cigarette smoking with corneal biomechanical properties and glaucoma-related traits, and to probe potential causal effects using Mendelian randomization (MR). Methods: Cross-sectional analyses within the UK Biobank (UKB) and Canadian Longitudinal Study on Aging (CLSA) cohorts. Multivariable linear and logistic regression models were used to assess associations of smoking (status, intensity, and duration) with corneal hysteresis (CH), corneal resistance factor, IOP, inner retinal thicknesses, and glaucoma. Two-sample MR analyses were performed. Results: Overall, 68,738 UKB (mean age, 56.7 years; 54.7% women) and 22 845 CLSA (mean age, 62.7 years; 49.1% women) participants were included. Compared with nonsmokers, smokers had a higher CH (UKB, +0.48 mm Hg; CLSA, +0.57 mm Hg; P < 0.001) and corneal resistance factor (UKB, +0.47 mm Hg; CLSA, +0.60 mm Hg; P < 0.001) with evidence of a dose-response effect in both studies. Differential associations with Goldmann-correlated IOP (UKB, +0.25 mm Hg; CLSA, +0.36 mm Hg; P < 0.001) and corneal-compensated IOP (UKB, -0.28 mm Hg; CLSA, -0.32 mm Hg; P ≤ 0.001) were observed. Smoking was not associated with inner retinal thicknesses or glaucoma status in either study. MR provided evidence for a causal effect of smoking on corneal biomechanics, especially higher CH. Conclusions: Cigarette smoking seems to increase corneal biomechanical resistance to deformation, but there was little evidence to support a relationship with glaucoma. This outcome may result in an artefactual association with measured IOP and could account for discordant results with glaucoma in previous epidemiological studies.

Determinants of non-attendance at face-to-face and telemedicine ophthalmic consultations.

BACKGROUND/AIMS: Evaluation of telemedicine care models has highlighted its potential for exacerbating healthcare inequalities. This study seeks to identify and characterise factors associated with non-attendance across face-to-face and telemedicine outpatient appointments. METHODS: A retrospective cohort study at a tertiary-level ophthalmic institution in the UK, between 1 January 2019 and 31 October 2021. Logistic regression modelled non-attendance against sociodemographic, clinical and operational exposure variables for all new patient registrations across five delivery modes: asynchronous, synchronous telephone, synchronous audiovisual and face to face prior to the pandemic and face to face during the pandemic. RESULTS: A total of 85 924 patients (median age 55 years, 54.4% female) were newly registered. Non-attendance differed significantly by delivery mode: (9.0% face to face prepandemic, 10.5% face to face during the pandemic, 11.7% asynchronous and 7.8%, synchronous during pandemic). Male sex, greater levels of deprivation, a previously cancelled appointment and not self-reporting ethnicity were strongly associated with non-attendance across all delivery modes. Individuals identifying as black ethnicity had worse attendance in synchronous audiovisual clinics (adjusted OR 4.24, 95% CI 1.59 to 11.28) but not asynchronous. Those not self-reporting their ethnicity were from more deprived backgrounds, had worse broadband access and had significantly higher non-attendance across all modes (all p<0.001). CONCLUSION: Persistent non-attendance among underserved populations attending telemedicine appointments highlights the challenge digital transformation faces for reducing healthcare inequalities. Implementation of new programmes should be accompanied by investigation into the differential health outcomes of vulnerable populations.

A foundation model for generalizable disease detection from retinal images.

Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders1. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications2. Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from unlabelled retinal images and provides a basis for label-efficient model adaptation in several applications. Specifically, RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels. We show that adapted RETFound consistently outperforms several comparison models in the diagnosis and prognosis of sight-threatening eye diseases, as well as incident prediction of complex systemic disorders such as heart failure and myocardial infarction with fewer labelled data. RETFound provides a generalizable solution to improve model performance and alleviate the annotation workload of experts to enable broad clinical AI applications from retinal imaging.

Differentiating glaucoma from chiasmal compression using optical coherence tomography: the macular naso-temporal ratio.

BACKGROUND/AIMS: The analysis of visual field loss patterns is clinically useful to guide differential diagnosis of visual pathway pathology. This study investigates whether a novel index of macular atrophy patterns can discriminate between chiasmal compression and glaucoma. METHODS: A retrospective series of patients with preoperative chiasmal compression, primary open-angle glaucoma (POAG) and healthy controls. Macular optical coherence tomography (OCT) images were analysed for the macular ganglion cell and inner plexiform layer (mGCIPL) thickness. The nasal hemi-macula was compared with the temporal hemi-macula to derive the macular naso-temporal ratio (mNTR). Differences between groups and diagnostic accuracy were explored with multivariable linear regression and the area under the receiver operating characteristic curve (AUC). RESULTS: We included 111 individuals (31 with chiasmal compression, 30 with POAG and 50 healthy controls). Compared with healthy controls, the mNTR was significantly greater in POAG cases (ß=0.07, 95% CI 0.03 to 0.11, p=0.001) and lower in chiasmal compression cases (ß=-0.12, 95% CI -0.16 to -0.09, p<0.001), even though overall mGCIPL thickness did not discriminate between these pathologies (p=0.36). The mNTR distinguished POAG from chiasmal compression with an AUC of 95.3% (95% CI 90% to 100%). The AUCs when comparing healthy controls to POAG and chiasmal compression were 79.0% (95% CI 68% to 90%) and 89.0% (95% CI 80% to 98%), respectively. CONCLUSIONS: The mNTR can distinguish between chiasmal compression and POAG with high discrimination. This ratio may provide utility over-and-above previously reported sectoral thinning metrics. Incorporation of mNTR into the output of OCT instruments may aid earlier diagnosis of chiasmal compression.

Development and international validation of custom-engineered and code-free deep-learning models for detection of plus disease in retinopathy of prematurity: a retrospective study.

BACKGROUND: Retinopathy of prematurity (ROP), a leading cause of childhood blindness, is diagnosed through interval screening by paediatric ophthalmologists. However, improved survival of premature neonates coupled with a scarcity of available experts has raised concerns about the sustainability of this approach. We aimed to develop bespoke and code-free deep learning-based classifiers for plus disease, a hallmark of ROP, in an ethnically diverse population in London, UK, and externally validate them in ethnically, geographically, and socioeconomically diverse populations in four countries and three continents. Code-free deep learning is not reliant on the availability of expertly trained data scientists, thus being of particular potential benefit for low resource health-care settings. METHODS: This retrospective cohort study used retinal images from 1370 neonates admitted to a neonatal unit at Homerton University Hospital NHS Foundation Trust, London, UK, between 2008 and 2018. Images were acquired using a Retcam Version 2 device (Natus Medical, Pleasanton, CA, USA) on all babies who were either born at less than 32 weeks gestational age or had a birthweight of less than 1501 g. Each images was graded by two junior ophthalmologists with disagreements adjudicated by a senior paediatric ophthalmologist. Bespoke and code-free deep learning models (CFDL) were developed for the discrimination of healthy, pre-plus disease, and plus disease. Performance was assessed internally on 200 images with the majority vote of three senior paediatric ophthalmologists as the reference standard. External validation was on 338 retinal images from four separate datasets from the USA, Brazil, and Egypt with images derived from Retcam and the 3nethra neo device (Forus Health, Bengaluru, India). FINDINGS: Of the 7414 retinal images in the original dataset, 6141 images were used in the final development dataset. For the discrimination of healthy versus pre-plus or plus disease, the bespoke model had an area under the curve (AUC) of 0·986 (95% CI 0·973-0·996) and the CFDL model had an AUC of 0·989 (0·979-0·997) on the internal test set. Both models generalised well to external validation test sets acquired using the Retcam for discriminating healthy from pre-plus or plus disease (bespoke range was 0·975-1·000 and CFDL range was 0·969-0·995). The CFDL model was inferior to the bespoke model on discriminating pre-plus disease from healthy or plus disease in the USA dataset (CFDL 0·808 [95% CI 0·671-0·909, bespoke 0·942 [0·892-0·982]], p=0·0070). Performance also reduced when tested on the 3nethra neo imaging device (CFDL 0·865 [0·742-0·965] and bespoke 0·891 [0·783-0·977]). INTERPRETATION: Both bespoke and CFDL models conferred similar performance to senior paediatric ophthalmologists for discriminating healthy retinal images from ones with features of pre-plus or plus disease; however, CFDL models might generalise less well when considering minority classes. Care should be taken when testing on data acquired using alternative imaging devices from that used for the development dataset. Our study justifies further validation of plus disease classifiers in ROP screening and supports a potential role for code-free approaches to help prevent blindness in vulnerable neonates. FUNDING: National Institute for Health Research Biomedical Research Centre based at Moorfields Eye Hospital NHS Foundation Trust and the University College London Institute of Ophthalmology. TRANSLATIONS: For the Portuguese and Arabic translations of the abstract see Supplementary Materials section.

Visual dysfunction is a better predictor than retinal thickness for dementia in Parkinson's disease.

BACKGROUND: Dementia is a common and devastating symptom of Parkinson's disease (PD). Visual function and retinal structure are both emerging as potentially predictive for dementia in Parkinson's but lack longitudinal evidence. METHODS: We prospectively examined higher order vision (skew tolerance and biological motion) and retinal thickness (spectral domain optical coherence tomography) in 100 people with PD and 29 controls, with longitudinal cognitive assessments at baseline, 18 months and 36 months. We examined whether visual and retinal baseline measures predicted longitudinal cognitive scores using linear mixed effects models and whether they predicted onset of dementia, death and frailty using time-to-outcome methods. RESULTS: Patients with PD with poorer baseline visual performance scored lower on a composite cognitive score (ß=0.178, SE=0.05, p=0.0005) and showed greater decreases in cognition over time (ß=0.024, SE=0.001, p=0.013). Poorer visual performance also predicted greater probability of dementia (χ² (1)=5.2, p=0.022) and poor outcomes (χ² (1) =10.0, p=0.002). Baseline retinal thickness of the ganglion cell-inner plexiform layer did not predict cognitive scores or change in cognition with time in PD (ß=-0.013, SE=0.080, p=0.87; ß=0.024, SE=0.001, p=0.12). CONCLUSIONS: In our deeply phenotyped longitudinal cohort, visual dysfunction predicted dementia and poor outcomes in PD. Conversely, retinal thickness had less power to predict dementia. This supports mechanistic models for Parkinson's dementia progression with onset in cortical structures and shows potential for visual tests to enable stratification for clinical trials.

Association Between Retinal Features From Multimodal Imaging and Schizophrenia.

Importance: The potential association of schizophrenia with distinct retinal changes is of clinical interest but has been challenging to investigate because of a lack of sufficiently large and detailed cohorts. Objective: To investigate the association between retinal biomarkers from multimodal imaging (oculomics) and schizophrenia in a large real-world population. Design, Setting, and Participants: This cross-sectional analysis used data from a retrospective cohort of 154 830 patients 40 years and older from the AlzEye study, which linked ophthalmic data with hospital admission data across England. Patients attended Moorfields Eye Hospital, a secondary care ophthalmic hospital with a principal central site, 4 district hubs, and 5 satellite clinics in and around London, United Kingdom, and had retinal imaging during the study period (January 2008 and April 2018). Data were analyzed from January 2022 to July 2022. Main Outcomes and Measures: Retinovascular and optic nerve indices were computed from color fundus photography. Macular retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (mGC-IPL) thicknesses were extracted from optical coherence tomography. Linear mixed-effects models were used to examine the association between schizophrenia and retinal biomarkers. Results: A total of 485 individuals (747 eyes) with schizophrenia (mean [SD] age, 64.9 years [12.2]; 258 [53.2%] female) and 100 931 individuals (165 400 eyes) without schizophrenia (mean age, 65.9 years [13.7]; 53 253 [52.8%] female) were included after images underwent quality control and potentially confounding conditions were excluded. Individuals with schizophrenia were more likely to have hypertension (407 [83.9%] vs 49 971 [48.0%]) and diabetes (364 [75.1%] vs 28 762 [27.6%]). The schizophrenia group had thinner mGC-IPL (-4.05 µm, 95% CI, -5.40 to -2.69; P = 5.4 × 10-9), which persisted when investigating only patients without diabetes (-3.99 µm; 95% CI, -6.67 to -1.30; P = .004) or just those 55 years and younger (-2.90 µm; 95% CI, -5.55 to -0.24; P = .03). On adjusted analysis, retinal fractal dimension among vascular variables was reduced in individuals with schizophrenia (-0.14 units; 95% CI, -0.22 to -0.05; P = .001), although this was not present when excluding patients with diabetes. Conclusions and Relevance: In this study, patients with schizophrenia had measurable differences in neural and vascular integrity of the retina. Differences in retinal vasculature were mostly secondary to the higher prevalence of diabetes and hypertension in patients with schizophrenia. The role of retinal features as adjunct outcomes in patients with schizophrenia warrants further investigation.

SynthEye: Investigating the Impact of Synthetic Data on Artificial Intelligence-assisted Gene Diagnosis of Inherited Retinal Disease.

Purpose: Rare disease diagnosis is challenging in medical image-based artificial intelligence due to a natural class imbalance in datasets, leading to biased prediction models. Inherited retinal diseases (IRDs) are a research domain that particularly faces this issue. This study investigates the applicability of synthetic data in improving artificial intelligence-enabled diagnosis of IRDs using generative adversarial networks (GANs). Design: Diagnostic study of gene-labeled fundus autofluorescence (FAF) IRD images using deep learning. Participants: Moorfields Eye Hospital (MEH) dataset of 15 692 FAF images obtained from 1800 patients with confirmed genetic diagnosis of 1 of 36 IRD genes. Methods: A StyleGAN2 model is trained on the IRD dataset to generate 512 × 512 resolution images. Convolutional neural networks are trained for classification using different synthetically augmented datasets, including real IRD images plus 1800 and 3600 synthetic images, and a fully rebalanced dataset. We also perform an experiment with only synthetic data. All models are compared against a baseline convolutional neural network trained only on real data. Main Outcome Measures: We evaluated synthetic data quality using a Visual Turing Test conducted with 4 ophthalmologists from MEH. Synthetic and real images were compared using feature space visualization, similarity analysis to detect memorized images, and Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) score for no-reference-based quality evaluation. Convolutional neural network diagnostic performance was determined on a held-out test set using the area under the receiver operating characteristic curve (AUROC) and Cohen's Kappa (κ). Results: An average true recognition rate of 63% and fake recognition rate of 47% was obtained from the Visual Turing Test. Thus, a considerable proportion of the synthetic images were classified as real by clinical experts. Similarity analysis showed that the synthetic images were not copies of the real images, indicating that copied real images, meaning the GAN was able to generalize. However, BRISQUE score analysis indicated that synthetic images were of significantly lower quality overall than real images (P < 0.05). Comparing the rebalanced model (RB) with the baseline (R), no significant change in the average AUROC and κ was found (R-AUROC = 0.86[0.85-88], RB-AUROC = 0.88[0.86-0.89], R-k = 0.51[0.49-0.53], and RB-k = 0.52[0.50-0.54]). The synthetic data trained model (S) achieved similar performance as the baseline (S-AUROC = 0.86[0.85-87], S-k = 0.48[0.46-0.50]). Conclusions: Synthetic generation of realistic IRD FAF images is feasible. Synthetic data augmentation does not deliver improvements in classification performance. However, synthetic data alone deliver a similar performance as real data, and hence may be useful as a proxy to real data. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references.

Phenotyping of ABCA4 Retinopathy by Machine Learning Analysis of Full-Field Electroretinography.

Purpose: Biallelic pathogenic variants in ABCA4 are the commonest cause of monogenic retinal disease. The full-field electroretinogram (ERG) quantifies severity of retinal dysfunction. We explored application of machine learning in ERG interpretation and in genotype-phenotype correlations. Methods: International standard ERGs in 597 cases of ABCA4 retinopathy were classified into three functional phenotypes by human experts: macular dysfunction alone (group 1), or with additional generalized cone dysfunction (group 2), or both cone and rod dysfunction (group 3). Algorithms were developed for automatic selection and measurement of ERG components and for classification of ERG phenotype. Elastic-net regression was used to quantify severity of specific ABCA4 variants based on effect on retinal function. Results: Of the cohort, 57.6%, 7.4%, and 35.0% fell into groups 1, 2, and 3 respectively. Compared with human experts, automated classification showed overall accuracy of 91.8% (SE, 0.169), and 96.7%, 39.3%, and 93.8% for groups 1, 2, and 3. When groups 2 and 3 were combined, the average holdout group accuracy was 93.6% (SE, 0.142). A regression model yielded phenotypic severity scores for the 47 commonest ABCA4 variants. Conclusions: This study quantifies prevalence of phenotypic groups based on retinal function in a uniquely large single-center cohort of patients with electrophysiologically characterized ABCA4 retinopathy and shows applicability of machine learning. Novel regression-based analyses of ABCA4 variant severity could identify individuals predisposed to severe disease. Translational Relevance: Machine learning can yield meaningful classifications of ERG data, and data-driven scoring of genetic variants can identify patients likely to benefit most from future therapies.

Diagnostic accuracy of code-free deep learning for detection and evaluation of posterior capsule opacification.

OBJECTIVE: To train and validate a code-free deep learning system (CFDLS) on classifying high-resolution digital retroillumination images of posterior capsule opacification (PCO) and to discriminate between clinically significant and non-significant PCOs. METHODS AND ANALYSIS: For this retrospective registry study, three expert observers graded two independent datasets of 279 images three separate times with no PCO to severe PCO, providing binary labels for clinical significance. The CFDLS was trained and internally validated using 179 images of a training dataset and externally validated with 100 images. Model development was through Google Cloud AutoML Vision. Intraobserver and interobserver variabilities were assessed using Fleiss kappa (κ) coefficients and model performance through sensitivity, specificity and area under the curve (AUC). RESULTS: Intraobserver variability κ values for observers 1, 2 and 3 were 0.90 (95% CI 0.86 to 0.95), 0.94 (95% CI 0.90 to 0.97) and 0.88 (95% CI 0.82 to 0.93). Interobserver agreement was high, ranging from 0.85 (95% CI 0.79 to 0.90) between observers 1 and 2 to 0.90 (95% CI 0.85 to 0.94) for observers 1 and 3. On internal validation, the AUC of the CFDLS was 0.99 (95% CI 0.92 to 1.0); sensitivity was 0.89 at a specificity of 1. On external validation, the AUC was 0.97 (95% CI 0.93 to 0.99); sensitivity was 0.84 and specificity was 0.92. CONCLUSION: This CFDLS provides highly accurate discrimination between clinically significant and non-significant PCO equivalent to human expert graders. The clinical value as a potential decision support tool in different models of care warrants further research.

AutoMorph: Automated Retinal Vascular Morphology Quantification Via a Deep Learning Pipeline.

Purpose: To externally validate a deep learning pipeline (AutoMorph) for automated analysis of retinal vascular morphology on fundus photographs. AutoMorph has been made publicly available, facilitating widespread research in ophthalmic and systemic diseases. Methods: AutoMorph consists of four functional modules: image preprocessing, image quality grading, anatomical segmentation (including binary vessel, artery/vein, and optic disc/cup segmentation), and vascular morphology feature measurement. Image quality grading and anatomical segmentation use the most recent deep learning techniques. We employ a model ensemble strategy to achieve robust results and analyze the prediction confidence to rectify false gradable cases in image quality grading. We externally validate the performance of each module on several independent publicly available datasets. Results: The EfficientNet-b4 architecture used in the image grading module achieves performance comparable to that of the state of the art for EyePACS-Q, with an F1-score of 0.86. The confidence analysis reduces the number of images incorrectly assessed as gradable by 76%. Binary vessel segmentation achieves an F1-score of 0.73 on AV-WIDE and 0.78 on DR HAGIS. Artery/vein scores are 0.66 on IOSTAR-AV, and disc segmentation achieves 0.94 in IDRID. Vascular morphology features measured from the AutoMorph segmentation map and expert annotation show good to excellent agreement. Conclusions: AutoMorph modules perform well even when external validation data show domain differences from training data (e.g., with different imaging devices). This fully automated pipeline can thus allow detailed, efficient, and comprehensive analysis of retinal vascular morphology on color fundus photographs. Translational Relevance: By making AutoMorph publicly available and open source, we hope to facilitate ophthalmic and systemic disease research, particularly in the emerging field of oculomics.

Evaluating an automated machine learning model that predicts visual acuity outcomes in patients with neovascular age-related macular degeneration.

PURPOSE: Neovascular age-related macular degeneration (nAMD) is a major global cause of blindness. Whilst anti-vascular endothelial growth factor (anti-VEGF) treatment is effective, response varies considerably between individuals. Thus, patients face substantial uncertainty regarding their future ability to perform daily tasks. In this study, we evaluate the performance of an automated machine learning (AutoML) model which predicts visual acuity (VA) outcomes in patients receiving treatment for nAMD, in comparison to a manually coded model built using the same dataset. Furthermore, we evaluate model performance across ethnic groups and analyse how the models reach their predictions. METHODS: Binary classification models were trained to predict whether patients' VA would be 'Above' or 'Below' a score of 70 one year after initiating treatment, measured using the Early Treatment Diabetic Retinopathy Study (ETDRS) chart. The AutoML model was built using the Google Cloud Platform, whilst the bespoke model was trained using an XGBoost framework. Models were compared and analysed using the What-if Tool (WIT), a novel model-agnostic interpretability tool. RESULTS: Our study included 1631 eyes from patients attending Moorfields Eye Hospital. The AutoML model (area under the curve [AUC], 0.849) achieved a highly similar performance to the XGBoost model (AUC, 0.847). Using the WIT, we found that the models over-predicted negative outcomes in Asian patients and performed worse in those with an ethnic category of Other. Baseline VA, age and ethnicity were the most important determinants of model predictions. Partial dependence plot analysis revealed a sigmoidal relationship between baseline VA and the probability of an outcome of 'Above'. CONCLUSION: We have described and validated an AutoML-WIT pipeline which enables clinicians with minimal coding skills to match the performance of a state-of-the-art algorithm and obtain explainable predictions.

The Risk of Sympathetic Ophthalmia after Vitreoretinal Surgery.

PURPOSE: To investigate the clinical course and outcomes of sympathetic ophthalmia (SO) and correlate these with the nature of the inciting event and the number of vitreoretinal (VR) procedures undergone by patients. DESIGN: A retrospective case review. SUBJECTS: All patients diagnosed with SO who had been treated or monitored at a single center over a 15-year period. METHODS: A search of the electronic patient record system at Moorfields Eye Hospital, London, over a 15-year period (between January 2000 and December 2015) was carried out using the search terms "sympathetic," "ophthalmia," and "ophthalmitis." Sixty-one patients with available records were identified, and data were collected from their complete electronic and paper records. MAIN OUTCOME MEASURES: The main outcome measures were best-corrected visual acuity at 1 year and at the end of follow-up and the number of VR surgical procedures preceding the diagnosis of SO. Data on patient age, sex, disease duration, ocular and systemic manifestations, ocular complications, retinal angiography, and treatment were also collected. RESULTS: There was a wide age range at presentation (2-84 years), and the length of follow-up ranged from 1 to 75 years. The first ocular event was trauma in 40 patients and surgery in 21 patients. Vitreoretinal surgery accounted for 13 of the 21 (62%) surgical first-event triggers. Twenty-three of 61 (38%) patients underwent VR surgery (1-7 operations) at some point before diagnosis. Surgical details were available for 15 patients, who had undergone a total of 25 VR procedures. Based on the surgical activity of the unit, the risk of developing SO after a single VR procedure was estimated to be 0.008%, rising to 6.67% with 7 procedures. A total of 23 (38%) patients experienced a decrease in acuity at the end of the follow-up period, vs. 9 (15%) patients experiencing an improvement and 18 (30%) remaining unchanged. CONCLUSIONS: We feel that the most significant finding in this study is the calculated risk of SO development after a single VR procedure, which was significantly lower in our cohort than that previously reported in the literature. This was seen to rise exponentially with additional procedures.