Relationship between macular perfusion and lesion distribution in diabetic retinopathy.

BACKGROUND/OBJECTIVES: To assess the relationship between macular vessel density metrics and foveal avascular zone (FAZ) characteristics on optical coherence tomography angiography (OCTA) and lesion distribution in eyes with diabetic retinopathy (DR). SUBJECTS/METHODS: Patients with DR who underwent both Optos ultrawidefield (UWF) pseudocolor imaging and macular OCTA (Cirrus Angioplex, 6 × 6 mm) were included in this cross-sectional observational study. The distribution of DR lesions was assessed by comparing each of the peripheral ETDRS extended fields (3-7) against their corresponding ETDRS field, hence eyes were defined as either having predominantly peripheral lesions (PPL) or predominantly central lesions (PCL). En face OCTA images from the superficial and deep capillary plexuses (SCP and DCP) were then analysed using Image J software. Perfusion density (PD), vessel length density (VLD), and fractal dimensions (FD) were calculated following binarization and skeletonization of the images. RESULTS: Out of 344 eyes, 116 (33.72%) eyes had PPL and 228 (66.28%) eyes had PCL. For all DRSS levels, VLD, PD, and FD were not significantly different between eyes with PPL and PCL. The FAZ in eyes with PPL, however, was found to be more circular in shape compared to eyes with PCL (p = 0.037). CONCLUSION: Although the presence of PPL has been associated with a higher risk for diabetic retinopathy progression, the macular perfusion is similar in eyes with PPL and PCL. The FAZ is more circular in eyes with PPL, but the clinical relevance of this difference remains to be defined.

Target Product Profile for a Machine Learning-Automated Retinal Imaging Analysis Software for Use in English Diabetic Eye Screening: Protocol for a Mixed Methods Study.

BACKGROUND: Diabetic eye screening (DES) represents a significant opportunity for the application of machine learning (ML) technologies, which may improve clinical and service outcomes. However, successful integration of ML into DES requires careful product development, evaluation, and implementation. Target product profiles (TPPs) summarize the requirements necessary for successful implementation so these can guide product development and evaluation. OBJECTIVE: This study aims to produce a TPP for an ML-automated retinal imaging analysis software (ML-ARIAS) system for use in DES in England. METHODS: This work will consist of 3 phases. Phase 1 will establish the characteristics to be addressed in the TPP. A list of candidate characteristics will be generated from the following sources: an overview of systematic reviews of diagnostic test TPPs; a systematic review of digital health TPPs; and the National Institute for Health and Care Excellence's Evidence Standards Framework for Digital Health Technologies. The list of characteristics will be refined and validated by a study advisory group (SAG) made up of representatives from key stakeholders in DES. This includes people with diabetes; health care professionals; health care managers and leaders; and regulators and policy makers. In phase 2, specifications for these characteristics will be drafted following a series of semistructured interviews with participants from these stakeholder groups. Data collected from these interviews will be analyzed using the shortlist of characteristics as a framework, after which specifications will be drafted to create a draft TPP. Following approval by the SAG, in phase 3, the draft will enter an internet-based Delphi consensus study with participants sought from the groups previously identified, as well as ML-ARIAS developers, to ensure feasibility. Participants will be invited to score characteristic and specification pairs on a scale from "definitely exclude" to "definitely include," and suggest edits. The document will be iterated between rounds based on participants' feedback. Feedback on the draft document will be sought from a group of ML-ARIAS developers before its final contents are agreed upon in an in-person consensus meeting. At this meeting, representatives from the stakeholder groups previously identified (minus ML-ARIAS developers, to avoid bias) will be presented with the Delphi results and feedback of the user group and asked to agree on the final contents by vote. RESULTS: Phase 1 was completed in November 2023. Phase 2 is underway and expected to finish in March 2024. Phase 3 is expected to be complete in July 2024. CONCLUSIONS: The multistakeholder development of a TPP for an ML-ARIAS for use in DES in England will help developers produce tools that serve the needs of patients, health care providers, and their staff. The TPP development process will also provide methods and a template to produce similar documents in other disease areas. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/50568.

Sight threatening diabetic retinopathy in patients with macular telangiectasia type 2.

PURPOSE: Although diabetes is highly prevalent in patients with MacTel, progression to severe non-proliferative (NPDR) and proliferative diabetic retinopathy (PDR) is rarely reported. We report multimodal imaging features of sight-threatening diabetic retinopathy (STDR) in eyes with macular telangiectasia type 2 (MacTel). METHODS: Retrospective case series of seven participants of the MacTel Study at the Moorfields Eye Hospital NHS Foundation Trust study site and one patient from the Institute of Retina and Vitreous of Londrina, Brazil. Sight threatening diabetic retinopathy was defined as severe NPDR, PDR or diabetic macular edema. RESULTS: We report imaging features of 16 eyes of eight patients (7/8, 87.5% female) with diagnoses of MacTel and type 2 diabetes mellitus with STDR. Mean (SD) age was 56 (8.3) years. Patients were followed-up for a mean time of 9.1 (4.7) years. A total of 10/16 (62.5%) eyes showed PDR and 2/16 (12.5%) eyes presented a macular epiretinal neovascularization. CONCLUSIONS: People with diabetes mellitus and MacTel may not be protected from STDR as previously reported. Although the two diseases rarely co-exist, regular monitoring for diabetic retinopathy progression is recommended according to baseline retinopathy severity grades in line with established international guidelines. The presence of MacTel may not modify extended screening intervals, but there is no current evidence. The limited case series in the literature support treatment for complications and should follow the standard of care for either condition. Due to dual pathology, reactivation may be difficult to diagnose on standard imaging and multimodal imaging is recommended.

Retinal vessel volume reference database derived from volume-rendered optical coherence tomography angiography.

Optical coherence tomography angiography (OCTA) enables three-dimensional reconstruction of the functional blood vessels in the retina. Therefore, it enables the quantification of 3D retinal vessel parameters such as surface area and vessel volume. In spite of the widespread use of OCTA, no representative volume-rendered vessel volume (VV) data are published to date. In this study, OCTA 3 × 3 mm macular cubes were processed with volume-rendering techniques to measure VV in 203 eyes from 107 healthy volunteers. Generalized linear models (GLM) were constructed to assess the impact of age, gender, visual acuity (VA), spherical equivalent (SE), and axial length (AL) on VV. Overall mean VV was 0.23 ± 0.05mm3. Age and axial length showed a negative correlation with VV. However, GLM model analysis found that AL exerted the most pronounced influence on VV. No statistically significant associations were identified between gender or between left and right eyes. This is the first study to assess 3D OCTA VV and its naturally occurring variations in a large series of healthy subjects. It offers novel insights into the characterization of normal retinal vascular anatomy in healthy individuals, contributing to a valuable reference for future research in this field.

Artificial intelligence-supported diabetic retinopathy screening in Tanzania: rationale and design of a randomised controlled trial.

INTRODUCTION: Globally, diabetic retinopathy (DR) is a major cause of blindness. Sub-Saharan Africa is projected to see the largest proportionate increase in the number of people living with diabetes over the next two decades. Screening for DR is recommended to prevent sight loss; however, in many low and middle-income countries, because of a lack of specialist eye care staff, current screening services for DR are not optimal. The use of artificial intelligence (AI) for DR screening, which automates the grading of retinal photographs and provides a point-of-screening result, offers an innovative potential solution to improve DR screening in Tanzania. METHODS AND ANALYSIS: We will test the hypothesis that AI-supported DR screening increases the proportion of persons with true referable DR who attend the central ophthalmology clinic following referral after screening in a single-masked, parallel group, individually randomised controlled trial. Participants (2364) will be randomised (1:1 ratio) to either AI-supported or the standard of care DR screening pathway. Participants allocated to the AI-supported screening pathway will receive their result followed by point-of-screening counselling immediately after retinal image capture. Participants in the standard of care arm will receive their result and counselling by phone once the retinal images have been graded in the usual way (typically after 2-4 weeks). The primary outcome is the proportion of persons with true referable DR attending the central ophthalmology clinic within 8 weeks of screening. Secondary outcomes, by trial arm, include the proportion of persons attending the central ophthalmology clinic out of all those referred, sensitivity and specificity, number of false positive referrals, acceptability and fidelity of AI-supported screening. ETHICS AND DISSEMINATION: The London School of Hygiene & Tropical Medicine, Kilimanjaro Christian Medical Centre and Tanzanian National Institute of Medical Research ethics committees have approved the trial. The results will be submitted to peer-reviewed journals for publication. TRIAL REGISTRATION NUMBER: ISRCTN18317152.

Developing a Continuous Severity Scale for Macular Telangiectasia Type 2 Using Deep Learning and Implications for Disease Grading.

PURPOSE: Deep learning (DL) models have achieved state-of-the-art medical diagnosis classification accuracy. Current models are limited by discrete diagnosis labels, but could yield more information with diagnosis in a continuous scale. We developed a novel continuous severity scaling system for macular telangiectasia (MacTel) type 2 by combining a DL classification model with uniform manifold approximation and projection (UMAP). DESIGN: We used a DL network to learn a feature representation of MacTel severity from discrete severity labels and applied UMAP to embed this feature representation into 2 dimensions, thereby creating a continuous MacTel severity scale. PARTICIPANTS: A total of 2003 OCT volumes were analyzed from 1089 MacTel Project participants. METHODS: We trained a multiview DL classifier using multiple B-scans from OCT volumes to learn a previously published discrete 7-step MacTel severity scale. The classifiers' last feature layer was extracted as input for UMAP, which embedded these features into a continuous 2-dimensional manifold. The DL classifier was assessed in terms of test accuracy. Rank correlation for the continuous UMAP scale against the previously published scale was calculated. Additionally, the UMAP scale was assessed in the κ agreement against 5 clinical experts on 100 pairs of patient volumes. For each pair of patient volumes, clinical experts were asked to select the volume with more severe MacTel disease and to compare them against the UMAP scale. MAIN OUTCOME MEASURES: Classification accuracy for the DL classifier and κ agreement versus clinical experts for UMAP. RESULTS: The multiview DL classifier achieved top 1 accuracy of 63.3% (186/294) on held-out test OCT volumes. The UMAP metric showed a clear continuous gradation of MacTel severity with a Spearman rank correlation of 0.84 with the previously published scale. Furthermore, the continuous UMAP metric achieved κ agreements of 0.56 to 0.63 with 5 clinical experts, which was comparable with interobserver κ values. CONCLUSIONS: Our UMAP embedding generated a continuous MacTel severity scale, without requiring continuous training labels. This technique can be applied to other diseases and may lead to more accurate diagnosis, improved understanding of disease progression, and key imaging features for pathologic characteristics. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Human selection bias drives the linear nature of the more ground truth effect in explainable deep learning optical coherence tomography image segmentation.

Supervised deep learning (DL) algorithms are highly dependent on training data for which human graders are assigned, for example, for optical coherence tomography (OCT) image annotation. Despite the tremendous success of DL, due to human judgment, these ground truth labels can be inaccurate and/or ambiguous and cause a human selection bias. We therefore investigated the impact of the size of the ground truth and variable numbers of graders on the predictive performance of the same DL architecture and repeated each experiment three times. The largest training dataset delivered a prediction performance close to that of human experts. All DL systems utilized were highly consistent. Nevertheless, the DL under-performers could not achieve any further autonomous improvement even after repeated training. Furthermore, a quantifiable linear relationship between ground truth ambiguity and the beneficial effect of having a larger amount of ground truth data was detected and marked as the more-ground-truth effect.

What are the perceptions and concerns of people living with diabetes and National Health Service staff around the potential implementation of AI-assisted screening for diabetic eye disease? Development and validation of a survey for use in a secondary care screening setting.

INTRODUCTION: The English National Health Service (NHS) Diabetic Eye Screening Programme (DESP) performs around 2.3 million eye screening appointments annually, generating approximately 13 million retinal images that are graded by humans for the presence or severity of diabetic retinopathy. Previous research has shown that automated retinal image analysis systems, including artificial intelligence (AI), can identify images with no disease from those with diabetic retinopathy as safely and effectively as human graders, and could significantly reduce the workload for human graders. Some algorithms can also determine the level of severity of the retinopathy with similar performance to humans. There is a need to examine perceptions and concerns surrounding AI-assisted eye-screening among people living with diabetes and NHS staff, if AI was to be introduced into the DESP, to identify factors that may influence acceptance of this technology. METHODS AND ANALYSIS: People living with diabetes and staff from the North East London (NEL) NHS DESP were invited to participate in two respective focus groups to codesign two online surveys exploring their perceptions and concerns around the potential introduction of AI-assisted screening.Focus group participants were representative of the local population in terms of ages and ethnicity. Participants' feedback was taken into consideration to update surveys which were circulated for further feedback. Surveys will be piloted at the NEL DESP and followed by semistructured interviews to assess accessibility, usability and to validate the surveys.Validated surveys will be distributed by other NHS DESP sites, and also via patient groups on social media, relevant charities and the British Association of Retinal Screeners. Post-survey evaluative interviews will be undertaken among those who consent to participate in further research. ETHICS AND DISSEMINATION: Ethical approval has been obtained by the NHS Research Ethics Committee (IRAS ID: 316631). Survey results will be shared and discussed with focus groups to facilitate preparation of findings for publication and to inform codesign of outreach activities to address concerns and perceptions identified.

Ethnic disparities in progression rates for sight-threatening diabetic retinopathy in diabetic eye screening: a population-based retrospective cohort study.

INTRODUCTION: The English Diabetic Eye Screening Programme (DESP) offers people living with diabetes (PLD) annual eye screening. We examined incidence and determinants of sight-threatening diabetic retinopathy (STDR) in a sociodemographically diverse multi-ethnic population. RESEARCH DESIGN AND METHODS: North East London DESP cohort data (January 2012 to December 2021) with 137 591 PLD with no retinopathy, or non-STDR at baseline in one/both eyes, were used to calculate STDR incidence rates by sociodemographic factors, diabetes type, and duration. HR from Cox models examined associations with STDR. RESULTS: There were 16 388 incident STDR cases over a median of 5.4 years (IQR 2.8-8.2; STDR rate 2.214, 95% CI 2.214 to 2.215 per 100 person-years). People with no retinopathy at baseline had a lower risk of sight-threatening diabetic retinopathy (STDR) compared with those with non-STDR in one eye (HR 3.03, 95% CI 2.91 to 3.15, p<0.001) and both eyes (HR 7.88, 95% CI 7.59 to 8.18, p<0.001). Black and South Asian individuals had higher STDR hazards than white individuals (HR 1.57, 95% CI 1.50 to 1.64 and HR 1.36, 95% CI 1.31 to 1.42, respectively). Additionally, every 5-year increase in age at inclusion was associated with an 8% reduction in STDR hazards (p<0.001). CONCLUSIONS: Ethnic disparities exist in a health system limited by capacity rather than patient economic circumstances. Diabetic retinopathy at first screen is a strong determinant of STDR development. By using basic demographic characteristics, screening programmes or clinical practices can stratify risk for sight-threatening diabetic retinopathy development.

Novel Outer Retinal Columnar Abnormalities (ORCA) and Non-Vasogenic Cystoid Macular Edema in Dense Deposit Disease.

PURPOSE: To describe the occurrence of bilateral outer retinal columnar abnormalities, non-vasogenic cystoid macular edema, and drusen in the context of dense deposit disease. METHODS: Case report. PATIENT: An 18-year-old female with dense deposit disease was referred to our specialist center for diagnosis and management with findings consistent with bilateral non-vasogenic cystoid macular edema and drusen. She was followed up in our clinic for forty months and treated with acetazolamide and ketorolac drops. RESULTS: Baseline examination revealed bilateral visual acuity (VA) reduction, and macular elevation with peripapillary drusen on fundus biomicroscopy. Optical coherence tomography revealed bilateral hyporeflective cystoid central macula changes, microcystoid changes with increased central subfield thickness (>450 microns), and outer retinal columnar abnormalities (ORCAs). Fluorescein angiography showed no evidence of macular leakage. Electrodiagnostic testing was within normal limits. Over the course of follow-up, she received treatment with acetazolamide 250mg BD PO and ketorolac 0.5% eye drops, with a partial reduction in her edema and improvement in VA. CONCLUSION: Dense deposit disease is a rare disease secondary to complement cascade dysregulation, associated with drusen. To the best of our knowledge, this is the first report of bilateral non-vasogenic cystoid macular edema and ORCA in a young female patient with dense deposit disease, confirmed with multimodal imaging.

Two-year recall for people with no diabetic retinopathy: a multi-ethnic population-based retrospective cohort study using real-world data to quantify the effect.

BACKGROUND/AIMS: The English Diabetic Eye Screening Programme (DESP) offers people living with diabetes (PLD) annual screening. Less frequent screening has been advocated among PLD without diabetic retinopathy (DR), but evidence for each ethnic group is limited. We examined the potential effect of biennial versus annual screening on the detection of sight-threatening diabetic retinopathy (STDR) and proliferative diabetic retinopathy (PDR) among PLD without DR from a large urban multi-ethnic English DESP. METHODS: PLD in North-East London DESP (January 2012 to December 2021) with no DR on two prior consecutive screening visits with up to 8 years of follow-up were examined. Annual STDR and PDR incidence rates, overall and by ethnicity, were quantified. Delays in identification of STDR and PDR events had 2-year screening intervals been used were determined. FINDINGS: Among 82 782 PLD (37% white, 36% South Asian, and 16% black people), there were 1788 incident STDR cases over mean (SD) 4.3 (2.4) years (STDR rate 0.51, 95% CI 0.47 to 0.55 per 100-person-years). STDR incidence rates per 100-person-years by ethnicity were 0.55 (95% CI 0.48 to 0.62) for South Asian, 0.34 (95% CI 0.29 to 0.40) for white, and 0.77 (95% CI 0.65 to 0.90) for black people. Biennial screening would have delayed diagnosis by 1 year for 56.3% (1007/1788) with STDR and 43.6% (45/103) with PDR. Standardised cumulative rates of delayed STDR per 100 000 persons for each ethnic group were 1904 (95% CI 1683 to 2154) for black people, 1276 (95% CI 1153 to 1412) for South Asian people, and 844 (95% CI 745 to 955) for white people. INTERPRETATION: Biennial screening would have delayed detection of some STDR and PDR by 1 year, especially among those of black ethnic origin, leading to healthcare inequalities.

Ethnicity is not biology: retinal pigment score to evaluate biological variability from ophthalmic imaging using machine learning.

Background: Few metrics exist to describe phenotypic diversity within ophthalmic imaging datasets, with researchers often using ethnicity as an inappropriate marker for biological variability. Methods: We derived a continuous, measured metric, the retinal pigment score (RPS), that quantifies the degree of pigmentation from a colour fundus photograph of the eye. RPS was validated using two large epidemiological studies with demographic and genetic data (UK Biobank and EPIC-Norfolk Study). Findings: A genome-wide association study (GWAS) of RPS from UK Biobank identified 20 loci with known associations with skin, iris and hair pigmentation, of which 8 were replicated in the EPIC-Norfolk cohort. There was a strong association between RPS and ethnicity, however, there was substantial overlap between each ethnicity and the respective distributions of RPS scores. Interpretation: RPS serves to decouple traditional demographic variables, such as ethnicity, from clinical imaging characteristics. RPS may serve as a useful metric to quantify the diversity of the training, validation, and testing datasets used in the development of AI algorithms to ensure adequate inclusion and explainability of the model performance, critical in evaluating all currently deployed AI models. The code to derive RPS is publicly available at: https://github.com/uw-biomedical-ml/retinal-pigmentation-score. Funding: The authors did not receive support from any organisation for the submitted work.

Progression, reliability, predicting parameters and sample size calculations for quantitative fundus autofluorescence measures in ABCA4-related retinopathy.

BACKGROUND/AIMS: To investigate the progression of quantitative autofluorescence (qAF) measures and the potential as clinical trial endpoint in ABCA4-related retinopathy. METHODS: In this longitudinal monocentre study, 64 patients with ABCA4-related retinopathy (age (mean±SD), 34.84±16.36 years) underwent serial retinal imaging, including optical coherence tomography (OCT) and qAF (488 nm excitation) imaging using a modified confocal scanning laser ophthalmoscope with a mean (±SD) review period of 20.32±10.90 months. A group of 110 healthy subjects served as controls. Retest variability, changes of qAF measures over time and its association with genotype and phenotype were analysed. Furthermore, individual prognostic feature importance was assessed, and sample size calculations for future interventional trials were performed. RESULTS: Compared with controls, qAF levels of patients were significantly elevated. The test-retest reliability revealed a 95% coefficient of repeatability of 20.37. During the observation time, young patients, patients with a mild phenotype (morphological and functional) and patients with mild mutations showed an absolute and relative increase in qAF values, while patients with advanced disease manifestation (morphological and functional), and homozygous mutations at adulthood revealed a decrease in qAF. Considering these parameters, required sample size and study duration could significantly be reduced. CONCLUSION: Under standardised settings with elaborated conditions towards operators and analysis to counterbalance variability, qAF imaging might be reliable, suitable for quantifying disease progression and constitutes a potential clinical surrogate marker in ABCA4-related retinopathy. Trial design based on patients' baseline characteristics and genotype has the potential to provide benefits regarding required cohort size and absolute number of visits.

Validation of collaborative cyberspace virtual reality oculometry enhanced with near real-time spatial audio.

Currently, most medical image data, such as optical coherence tomography (OCT) images, are displayed in two dimensions on a computer screen. Advances in computer information technology have contributed to the growing storage of these data in electronic form. However, the data are usually processed only locally on site. To overcome such hurdles, a cyberspace virtual reality (csVR) application was validated, in which interactive OCT data were presented simultaneously to geographically distant sites (Lucerne, London, and Barcelona) where three graders independently measured the ocular csVR OCT diameters. A total of 109 objects were measured, each three times, resulting in a total of 327 csVR measurements. A minor mean absolute difference of 5.3 µm was found among the 3 measurements of an object (standard deviation 4.2 µm, coefficient of variation 0.3% with respect to the mean object size). Despite the 5 h of online work, csVR was well tolerated and safe. Digital high-resolution OCT data can be remotely and collaboratively processed in csVR. With csVR, measurements and actions enhanced with spatial audio communication can be made consistently in near real time, even if the users are situated geographically far apart. The proposed visuo-auditory framework has the potential to further boost the convenience of digital medicine toward csVR precision and collaborative medicine.

HISTOPATHOLOGIC CONFIRMATION OF LYMPHOCYTIC INFILTRATION OF THE OPTIC NERVE AND INNER RETINA IN BIRDSHOT CHORIORETINOPATHY.

PURPOSE: To describe new histological findings involving the inner retina in birdshot chorioretinopathy. METHODS: Evaluation of the inner retinal pathology of the eye of a patient with bilateral birdshot chorioretinopathy who underwent enucleation for a unilateral ciliochoroidal melanoma. RESULTS: Histopathological sections showed focal perivascular lymphocytic infiltration at the optic nerve head that extended into the adjacent inner retina, mainly involving the ganglion and nerve fiber layers. CONCLUSION: We have previously shown that birdshot chorioretinopathy has multiple foci of lymphocytes in the choroid. This is the first report that demonstrates lymphocytic infiltration of the inner retinal layers. This may lead to the bipolar and Müller cell dysfunction that ultimately results in an electronegative electroretinogram.

Formal registration of visual impairment in people with diabetic retinopathy significantly underestimates the scale of the problem: a retrospective cohort study at a tertiary care eye hospital service in the UK.

AIMS: To analyse the prevalence of visual impairment (VI), compare it to certification of visual impairment (CVI) and analyse VI associations in patients with diabetic retinopathy (DR). METHODS: Retrospective cohort study, which included 8007 patients with DR referred from the English diabetic eye screening programme to a tertiary referral eye hospital. Main outcome measure was VI, defined as vision in the best eye of <6/24. We conducted a multivariable logistic regression for VI as primary outcome of interest, controlling for age, sex, type of diabetes, baseline DR grade, ethnicity and index of multiple deprivation (IMD). RESULTS: Mean age was 64.5 (SD 13.6) years; 61% of patients were men; and 31% of South Asian ethnicity. There were 68 patients with CVI during the study period, and 84% (272/325) of patients with VI did not have CVI after a mean follow-up of 1.87 (SD ±0.86) years. Older age showed a positive association with VI (OR per decade rise 1.88, 95% CI 1.70 to 2.08; p=1.8×10-34). Men had a lower risk of VI (OR 0.62, 95% CI 0.50 to 0.79, p=6.0×10-5), and less deprivation had a graded inverse association with VI (OR per IMD category increase 0.83, 95% CI 0.74 to 0.93, p value for linear trend 0.002). CONCLUSION: The majority of people with vision impairment are not registered at the point of care, which could translate to underestimation of diabetes-related VI and all-cause VI at a national level if replicated at other centres. Further work is needed to explore rates of VI and uptake of registration.

Training Deep Learning Models to Work on Multiple Devices by Cross-Domain Learning with No Additional Annotations.

PURPOSE: To create an unsupervised cross-domain segmentation algorithm for segmenting intraretinal fluid and retinal layers on normal and pathologic macular OCT images from different manufacturers and camera devices. DESIGN: We sought to use generative adversarial networks (GANs) to generalize a segmentation model trained on one OCT device to segment B-scans obtained from a different OCT device manufacturer in a fully unsupervised approach without labeled data from the latter manufacturer. PARTICIPANTS: A total of 732 OCT B-scans from 4 different OCT devices (Heidelberg Spectralis, Topcon 1000, Maestro2, and Zeiss Plex Elite 9000). METHODS: We developed an unsupervised GAN model, GANSeg, to segment 7 retinal layers and intraretinal fluid in Topcon 1000 OCT images (domain B) that had access only to labeled data on Heidelberg Spectralis images (domain A). GANSeg was unsupervised because it had access only to 110 Heidelberg labeled OCTs and 556 raw and unlabeled Topcon 1000 OCTs. To validate GANSeg segmentations, 3 masked graders manually segmented 60 OCTs from an external Topcon 1000 test dataset independently. To test the limits of GANSeg, graders also manually segmented 3 OCTs from Zeiss Plex Elite 9000 and Topcon Maestro2. A U-Net was trained on the same labeled Heidelberg images as baseline. The GANSeg repository with labeled annotations is at https://github.com/uw-biomedical-ml/ganseg. MAIN OUTCOME MEASURES: Dice scores comparing segmentation results from GANSeg and the U-Net model with the manual segmented images. RESULTS: Although GANSeg and U-Net achieved comparable Dice scores performance as human experts on the labeled Heidelberg test dataset, only GANSeg achieved comparable Dice scores with the best performance for the ganglion cell layer plus inner plexiform layer (90%; 95% confidence interval [CI], 68%-96%) and the worst performance for intraretinal fluid (58%; 95% CI, 18%-89%), which was statistically similar to human graders (79%; 95% CI, 43%-94%). GANSeg significantly outperformed the U-Net model. Moreover, GANSeg generalized to both Zeiss and Topcon Maestro2 swept-source OCT domains, which it had never encountered before. CONCLUSIONS: GANSeg enables the transfer of supervised deep learning algorithms across OCT devices without labeled data, thereby greatly expanding the applicability of deep learning algorithms.

A Deep Learning Framework for the Detection and Quantification of Reticular Pseudodrusen and Drusen on Optical Coherence Tomography.

Purpose: The purpose of this study was to develop and validate a deep learning (DL) framework for the detection and quantification of reticular pseudodrusen (RPD) and drusen on optical coherence tomography (OCT) scans. Methods: A DL framework was developed consisting of a classification model and an out-of-distribution (OOD) detection model for the identification of ungradable scans; a classification model to identify scans with drusen or RPD; and an image segmentation model to independently segment lesions as RPD or drusen. Data were obtained from 1284 participants in the UK Biobank (UKBB) with a self-reported diagnosis of age-related macular degeneration (AMD) and 250 UKBB controls. Drusen and RPD were manually delineated by five retina specialists. The main outcome measures were sensitivity, specificity, area under the receiver operating characteristic (ROC) curve (AUC), kappa, accuracy, intraclass correlation coefficient (ICC), and free-response receiver operating characteristic (FROC) curves. Results: The classification models performed strongly at their respective tasks (0.95, 0.93, and 0.99 AUC, respectively, for the ungradable scans classifier, the OOD model, and the drusen and RPD classification models). The mean ICC for the drusen and RPD area versus graders was 0.74 and 0.61, respectively, compared with 0.69 and 0.68 for intergrader agreement. FROC curves showed that the model's sensitivity was close to human performance. Conclusions: The models achieved high classification and segmentation performance, similar to human performance. Translational Relevance: Application of this robust framework will further our understanding of RPD as a separate entity from drusen in both research and clinical settings.

Properties of Patient-Reported Outcome Measures in Recessive Stargardt Disease.

INTRODUCTION: The study aimed to explore the psychometric properties of the National Eye Institute Visual Function Questionnaire (NEI VFQ) and Impact of Vision Impairment (IVI) profile in recessive Stargardt disease (STGD1). METHODS: The NEI VFQ-25 and IVI-28 were administered to individuals with STGD1. Responses were analyzed following psychometrically established dimension structures of the NEI VFQ (visual function [VF] subscale; socioemotional [SE] subscale) and of the IVI (functional [F] subscale; emotional [E] subscale). We analyzed internal consistency, dimensionality, item fit, and differential item functioning (DIF), using latent trait models. Criterion validity was assessed using Pearson correlation coefficients. RESULTS: Seventy-one participants (42 females, 29 males; mean age, 44 ± 19 years) were included. Self-reported difficulty levels were lower than the mean difficulty of items in both instruments. Person reliability and person separation index of the instruments were 0.85 and 2.40 (NEI VFQ-VF), 0.69 and 1.49 (NEI-VFQ-SE), 0.88 and 2.77 (IVI-F), and 0.72 and 1.62 (IVI-E). No items showed misfit at a level distorting the measurement system. One IVI item showed DIF by gender but was retained as person measures were largely unaffected by its removal. NEI VFQ-VF and IVI-F as well as NEI VFQ-SE and IVI-E were positively correlated (r = 0.79 and 0.64, respectively). CONCLUSION: The NEI VFQ and IVI have acceptable psychometric properties in STGD1 with the IVI allowing more sensitive person stratification. Targeting of questionnaires to individuals with STGD1 might be improved by including additional content domains specific to the disease.