Retinopathy Associated With Coxsackie B Virus Infection.

This case report describes a diagnosis of unilateral retinopathy secondary to coxsackie B virus in a male patient aged 41 years who presented with a central scotoma and blistering rash of the hands, feet, and mouth for 4 days.

Implementation of optical coherence tomography in retinopathy of prematurity screening.

PURPOSE OF REVIEW: In this review, we explore the investigational applications of optical coherence tomography (OCT) in retinopathy of prematurity (ROP), the insights they have delivered thus far, and key milestones for its integration into the standard of care. RECENT FINDINGS: While OCT has been widely integrated into clinical management of common retinal diseases, its use in pediatric contexts has been undermined by limitations in ergonomics, image acquisition time, and field of view. Recently, investigational handheld OCT devices have been reported with advancements including ultra-widefield view, noncontact use, and high-speed image capture permitting real-time en face visualization. These developments are compelling for OCT as a more objective alternative with reduced neonatal stress compared to indirect ophthalmoscopy and/or fundus photography as a means of classifying and monitoring ROP. SUMMARY: OCT may become a viable modality in management of ROP. Ongoing innovation surrounding handheld devices should aim to optimize patient comfort and image resolution in the retinal periphery. Future clinical investigations may seek to objectively characterize features of peripheral stage and explore novel biomarkers of disease activity.

Assessing the Impact of Image Quality on Deep Learning Classification of Infectious Keratitis.

Objective: To investigate the impact of corneal photograph quality on convolutional neural network (CNN) predictions. Design: A CNN trained to classify bacterial and fungal keratitis was evaluated using photographs of ulcers labeled according to 5 corneal image quality parameters: eccentric gaze direction, abnormal eyelid position, over/under-exposure, inadequate focus, and malpositioned light reflection. Participants: All eligible subjects with culture and stain-proven bacterial and/or fungal ulcers presenting to Aravind Eye Hospital in Madurai, India, between January 1, 2021 and December 31, 2021. Methods: Convolutional neural network classification performance was compared for each quality parameter, and gradient class activation heatmaps were generated to visualize regions of highest influence on CNN predictions. Main Outcome Measures: Area under the receiver operating characteristic and precision recall curves were calculated to quantify model performance. Bootstrapped confidence intervals were used for statistical comparisons. Logistic loss was calculated to measure individual prediction accuracy. Results: Individual presence of either light reflection or eyelids obscuring the corneal surface was associated with significantly higher CNN performance. No other quality parameter significantly influenced CNN performance. Qualitative review of gradient class activation heatmaps generally revealed the infiltrate as having the highest diagnostic relevance. Conclusions: The CNN demonstrated expert-level performance regardless of image quality. Future studies may investigate use of smartphone cameras and image sets with greater variance in image quality to further explore the influence of these parameters on model performance. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

Improved Training Efficiency for Retinopathy of Prematurity Deep Learning Models Using Comparison versus Class Labels.

Purpose: To compare the efficacy and efficiency of training neural networks for medical image classification using comparison labels indicating relative disease severity versus diagnostic class labels from a retinopathy of prematurity (ROP) image dataset. Design: Evaluation of diagnostic test or technology. Participants: Deep learning neural networks trained on expert-labeled wide-angle retinal images obtained from patients undergoing diagnostic ROP examinations obtained as part of the Imaging and Informatics in ROP (i-ROP) cohort study. Methods: Neural networks were trained with either class or comparison labels indicating plus disease severity in ROP retinal fundus images from 2 datasets. After training and validation, all networks underwent evaluation using a separate test dataset in 1 of 2 binary classification tasks: normal versus abnormal or plus versus nonplus. Main Outcome Measures: Area under the receiver operating characteristic curve (AUC) values were measured to assess network performance. Results: Given the same number of labels, neural networks learned more efficiently by comparison, generating significantly higher AUCs in both classification tasks across both datasets. Similarly, given the same number of images, comparison learning developed networks with significantly higher AUCs across both classification tasks in 1 of 2 datasets. The difference in efficiency and accuracy between models trained on either label type decreased as the size of the training set increased. Conclusions: Comparison labels individually are more informative and more abundant per sample than class labels. These findings indicate a potential means of overcoming the common obstacle of data variability and scarcity when training neural networks for medical image classification tasks.

Federated Learning for Multicenter Collaboration in Ophthalmology: Implications for Clinical Diagnosis and Disease Epidemiology.

OBJECTIVE: To utilize a deep learning (DL) model trained via federated learning (FL), a method of collaborative training without sharing patient data, to delineate institutional differences in clinician diagnostic paradigms and disease epidemiology in retinopathy of prematurity (ROP). DESIGN: Evaluation of a diagnostic test or technology. SUBJECTS AND CONTROLS: We included 5245 patients with wide-angle retinal imaging from the neonatal intensive care units of 7 institutions as part of the Imaging and Informatics in ROP study. Images were labeled with the clinical diagnoses of plus disease (plus, preplus, no plus), which were documented in the chart, and a reference standard diagnosis was determined by 3 image-based ROP graders and the clinical diagnosis. METHODS: Demographics (birth weight, gestational age) and clinical diagnoses for all eye examinations were recorded from each institution. Using an FL approach, a DL model for plus disease classification was trained using only the clinical labels. The 3 class probabilities were then converted into a vascular severity score (VSS) for each eye examination, as well as an "institutional VSS," in which the average of the VSS values assigned to patients' higher severity ("worse") eyes at each examination was calculated for each institution. MAIN OUTCOME MEASURES: We compared demographics, clinical diagnoses of plus disease, and institutional VSSs between institutions using the McNemar-Bowker test, 2-proportion Z test, and 1-way analysis of variance with post hoc analysis by the Tukey-Kramer test. Single regression analysis was performed to explore the relationship between demographics and VSSs. RESULTS: We found that the proportion of patients diagnosed with preplus disease varied significantly between institutions (P < 0.001). Using the DL-derived VSS trained on the data from all institutions using FL, we observed differences in the institutional VSS and the level of vascular severity diagnosed as no plus (P < 0.001) across institutions. A significant, inverse relationship between the institutional VSS and mean gestational age was found (P = 0.049, adjusted R2 = 0.49). CONCLUSIONS: A DL-derived ROP VSS developed without sharing data between institutions using FL identified differences in the clinical diagnoses of plus disease and overall levels of ROP severity between institutions. Federated learning may represent a method to standardize clinical diagnoses and provide objective measurements of disease for image-based diseases.

Federated Learning for Multicenter Collaboration in Ophthalmology: Improving Classification Performance in Retinopathy of Prematurity.

OBJECTIVE: To compare the performance of deep learning classifiers for the diagnosis of plus disease in retinopathy of prematurity (ROP) trained using 2 methods for developing models on multi-institutional data sets: centralizing data versus federated learning (FL) in which no data leave each institution. DESIGN: Evaluation of a diagnostic test or technology. SUBJECTS: Deep learning models were trained, validated, and tested on 5255 wide-angle retinal images in the neonatal intensive care units of 7 institutions as part of the Imaging and Informatics in ROP study. All images were labeled for the presence of plus, preplus, or no plus disease with a clinical label and a reference standard diagnosis (RSD) determined by 3 image-based ROP graders and the clinical diagnosis. METHODS: We compared the area under the receiver operating characteristic curve (AUROC) for models developed on multi-institutional data, using a central approach initially, followed by FL, and compared locally trained models with both approaches. We compared the model performance (κ) with the label agreement (between clinical and RSD), data set size, and number of plus disease cases in each training cohort using the Spearman correlation coefficient (CC). MAIN OUTCOME MEASURES: Model performance using AUROC and linearly weighted κ. RESULTS: Four settings of experiment were used: FL trained on RSD against central trained on RSD, FL trained on clinical labels against central trained on clinical labels, FL trained on RSD against central trained on clinical labels, and FL trained on clinical labels against central trained on RSD (P = 0.046, P = 0.126, P = 0.224, and P = 0.0173, respectively). Four of the 7 (57%) models trained on local institutional data performed inferiorly to the FL models. The model performance for local models was positively correlated with the label agreement (between clinical and RSD labels, CC = 0.389, P = 0.387), total number of plus cases (CC = 0.759, P = 0.047), and overall training set size (CC = 0.924, P = 0.002). CONCLUSIONS: We found that a trained FL model performs comparably to a centralized model, confirming that FL may provide an effective, more feasible solution for interinstitutional learning. Smaller institutions benefit more from collaboration than larger institutions, showing the potential of FL for addressing disparities in resource access.

PROLIFERATIVE RETINOPATHY IN A 13-YEAR-OLD WITH ADAMS-OLIVER SYNDROME.

PURPOSE: Adams-Oliver syndrome is a rare, inherited disorder of embryologic development that affects multiple systems. Ocular manifestations have been poorly characterized because of the low prevalence and high mortality of the disease when it is associated with internal organ and/or ophthalmic manifestations. We present a case of Adams-Oliver syndrome in a 13-year-old patient whose multimodal retinal imaging findings helped direct management. METHODS: Single patient case report reviewing medical records and imaging. RESULTS: Visual acuity upon presentation was 20/40 in each eye. Ultra-widefield fluorescein angiography revealed peripheral nonperfusion with terminal vascular bulbs, and leakage from a temporal fibrovascular complex in the left eye. Fundus autofluorescence imaging showed hyperautofluorescence associated with optic disc drusen and the fibrovascular complex. Treatment with targeted laser photocoagulation was associated with regression of the neovascularization. CONCLUSION: Retinal manifestations of Adams-Oliver syndrome as observed with ultra-widefield fundus imaging may resemble those of familial exudative vitreoretinopathy and retinopathy of prematurity. Treatment of avascular retina with panretinal photocoagulation can be considered.

Pentosan polysulfate maculopathy.

Pentosan polysulfate sodium (PPS), a semisynthetic sulfated polysaccharide, is the only FDA-approved oral therapy for interstitial cystitis. Recent studies have described a progressive, vision-threatening macular condition associated with long-term PPS use. We reviewed all publications concerning PPS maculopathy to consolidate known clinical features and to evaluate the strength of this association. Current literature supports a strong dose-dependent association between PPS exposure and a progressive maculopathy impacting the retinal pigment epithelium (RPE) and RPE-photoreceptor interface that may worsen even after drug cessation. Initial symptoms may include prolonged dark adaptation and difficulty reading with relative visual acuity preservation. Fundus examination often shows macular pigment clumps corresponding to lesions of focal RPE thickening. Fundus autofluorescence most clearly depicts the condition, with a distinctive pattern of hypo- and hyperautofluorescent spots in the posterior pole that sometimes extends to the retinal periphery. Many cases also show a characteristic peripapillary hypoautofluorescent halo. Near infrared reflectance may aid in early detection. RPE atrophy, cystoid macular edema, and macular neovascularization may also occur, potentially resulting in loss of central acuity. This newly described association implies significant public health risk. Ophthalmologists should screen PPS users with multimodal retinal imaging, and prescribers should minimize dose and duration of PPS use.

Prevalence of persistent avascular retina in untreated children with a history of retinopathy of prematurity screening.

Persistent avascular retina (PAR) in prematurely born individuals may be a risk factor for late sequelae of retinopathy of prematurity (ROP), including retinal detachment in older childhood and adulthood. Although PAR has been associated with use of vascular endothelial growth factor antagonist therapy for treatment-requiring ROP, the prevalence of this finding in patients without prior ROP treatment is unknown. We performed a cross-sectional study to determine the prevalence of PAR in a cohort of patients 4-8 years of age who were screened for ROP in the neonatal intensive care unit and did not receive treatment. Patients were recruited from an existing population-based cohort and underwent ultra-widefield fluorescein angiography (UWFFA). UWFFA images of 43 eyes of 23 patients were evaluated. Average age at time of evaluation was 6.2 years. PAR was observed in 21 patients (91%). Thirteen eyes (30%) had PAR in zone II; 23 (53%), in zone III. Six eyes (14%) had abnormal vessels without clear PAR. These findings indicate a high prevalence of PAR in patients with a history of ROP screening without treatment.

Tauroursodeoxycholic Acid Protects Retinal and Visual Function in a Mouse Model of Type 1 Diabetes.

PURPOSE: Previous studies demonstrated that systemic treatment with tauroursodeoxycholic acid (TUDCA) is protective in in vivo mouse models of retinal degeneration and in culture models of hyperglycemia. This study tested the hypothesis that TUDCA will preserve visual and retinal function in a mouse model of early diabetic retinopathy (DR). METHODS: Adult C57BL/6J mice were treated with streptozotocin (STZ) and made diabetic at 8-10 weeks of age. Control and diabetic mice were treated with vehicle or TUDCA starting 1 or 3 weeks after induction of diabetes, and were assessed bimonthly for visual function via an optomotor response and monthly for retinal function via scotopic electroretinograms. RESULTS: Diabetic mice showed significantly reduced spatial frequency and contrast sensitivity thresholds compared to control mice, while diabetic mice treated early with TUDCA showed preservation at all timepoints. A-wave, b-wave, and oscillatory potential 2 (OP2) amplitudes decreased in diabetic mice. Diabetic mice also exhibited delays in a-wave and OP2-implicit times. Early TUDCA treatment ameliorated a-wave, b-wave, and OP2 deficits. Late TUDCA treatment showed reduced preservation effects compared to early treatment. CONCLUSIONS: Early TUDCA treatment preserved visual function in an STZ-mouse model of Type I diabetes. These data add to a growing body of preclinical research that may support testing whether TUDCA may be an effective early clinical intervention against declining visual function caused by diabetic retinopathy.

Applications of interpretability in deep learning models for ophthalmology.

PURPOSE OF REVIEW: In this article, we introduce the concept of model interpretability, review its applications in deep learning models for clinical ophthalmology, and discuss its role in the integration of artificial intelligence in healthcare. RECENT FINDINGS: The advent of deep learning in medicine has introduced models with remarkable accuracy. However, the inherent complexity of these models undermines its users' ability to understand, debug and ultimately trust them in clinical practice. Novel methods are being increasingly explored to improve models' 'interpretability' and draw clearer associations between their outputs and features in the input dataset. In the field of ophthalmology, interpretability methods have enabled users to make informed adjustments, identify clinically relevant imaging patterns, and predict outcomes in deep learning models. SUMMARY: Interpretability methods support the transparency necessary to implement, operate and modify complex deep learning models. These benefits are becoming increasingly demonstrated in models for clinical ophthalmology. As quality standards for deep learning models used in healthcare continue to evolve, interpretability methods may prove influential in their path to regulatory approval and acceptance in clinical practice.

Pentosan Polysulfate Maculopathy versus Inherited Macular Dystrophies: Comparative Assessment with Multimodal Imaging.

PURPOSE: To evaluate whether pentosan polysulfate (PPS) maculopathy manifests distinctive characteristics that permit differentiation from hereditary maculopathies with multimodal fundus imaging. DESIGN: Retrospective review. PARTICIPANTS: Emory Eye Center databases were queried for the following International Classification of Diseases codes from May 20, 2014, through October 22, 2019: 362.70 (unspecified hereditary retinal dystrophy), 362.74 + H35.52 (pigmentary retinal dystrophy), 362.76 + H35.54 (dystrophies primarily involving the retinal pigment epithelium), and H35.50 (unspecified macular degeneration). METHODS: Fundus images for each patient were evaluated, including color fundus photographs, fundus autofluorescence images, and spectral-domain OCT images. Cases with imaging sufficient for diagnostic classification were analyzed. Masked graders classified patient images as follows: highly suggestive of PPS maculopathy; some features resembling PPS maculopathy, but not classic disease; and clearly distinct from PPS maculopathy. MAIN OUTCOME MEASURES: Sensitivity and specificity for identification of PPS maculopathy by masked reviewers. RESULTS: A total of 1394 patients were evaluated, and 1131 had imaging sufficient for classification. Fifteen patients were categorized as having findings highly suggestive of PPS maculopathy; 25 patients showed some features resembling PPS maculopathy but not classic disease; and 1091 patients showed evidence of disease clearly distinct from PPS maculopathy. All 10 patients with PPS maculopathy in this dataset were correctly categorized as having PPS maculopathy. Five patients without PPS exposure were categorized incorrectly as having PPS maculopathy. This represented a 100% sensitivity and 99.6% specificity for identification of PPS maculopathy by masked review of fundus imaging in this dataset. CONCLUSIONS: The imaging characteristics of PPS maculopathy allow for differentiation from hereditary maculopathies even in the absence of known exposure to the drug.

Phenotypic Spectrum of Pentosan Polysulfate Sodium-Associated Maculopathy: A Multicenter Study.

IMPORTANCE: A unique pigmentary maculopathy was recently described in 6 patients with long-term exposure to pentosan polysulfate sodium (PPS), a long-standing oral therapy for interstitial cystitis. OBJECTIVE: To characterize the exposure characteristics and clinical manifestations of PPS-associated maculopathy. DESIGN, SETTING, AND PARTICIPANTS: In this multi-institutional case series, medical records of patients who exhibited the characteristic maculopathy in the setting of prior PPS exposure were retrospectively reviewed. Data were collected from August 1, 2012, to October 1, 2018, and data were analyzed from October 2018 to January 2019. MAIN OUTCOMES AND MEASURES: Drug exposure, visual acuity, and retinal imaging characteristics. RESULTS: Of the 35 included patients (70 eyes), 34 (97%) were female, and the median (range) age was 60 (37-79) years. The median (range) duration of PPS intake was 15 (3-22) years, and the median (range) cumulative exposure was 1.61 (0.44-4.31) kg. The leading visual symptoms were metamorphopsia, blurred vision, and prolonged dark adaptation. Median (range) logMAR visual acuity of all eyes was 0.10 (-0.12 to 1.18). Fundus examination often revealed hyperpigmented macular spots (34 of 64 eyes [53%]) with interspersed pale-yellow deposits, although less commonly in eyes that exhibited retinal pigment epithelial atrophy (6 of 26 eyes [23%]; P < .001). Optical coherence tomography showed foci of retinal pigment epithelium elevation or thickening associated with hyperreflectance on near-infrared reflectance imaging. Fundus autofluorescence imaging typically revealed a symmetric, confluent pattern of hyperautofluorescent and hypoautofluorescent spots that involved the fovea in all eyes and extended to the retinal periphery in 24 eyes (36%). Longitudinal evaluation demonstrated dynamic changes in pigmentary abnormalities. CONCLUSIONS AND RELEVANCE: These findings suggest that PPS-associated maculopathy is a vision-threatening condition that can manifest in the setting of long-term exposure to the drug. Multimodal imaging posits a distinctive clinical phenotype, characterized in this cohort by dynamic alterations within the retinal pigment epithelium and at the retinal pigment epithelium-photoreceptor interface. Ongoing work might explore causality and direct screening guidelines.

Late-Stage Sorsby Fundus Dystrophy Manifesting Severe Vision Loss in the Absence of Choroidal Neovascularization.

A patient with a family history of molecularly confirmed Sorsby fundus dystrophy (SFD) presented with 9 years of progressive, bilateral central vision loss. Specific mutation analysis of the TIMP3 gene confirmed SFD, identifying a pathogenic mutation of p.Ser204Cys:c.610A>T. Optical coherence tomography imaging revealed diffuse retinal, retinal pigment epithelium, and choroidal atrophy without evidence for choroidal neovascularization (CNV). Although SFD is classically associated with CNV and subretinal fibrosis, some cases follow an atrophic course in the absence of CNV formation. This case highlights the extent to which extensive atrophic degeneration can lead to visual disability without choroidal neovascularization in late-stage SFD. [Ophthalmic Surg Lasers Imaging Retina. 2019;50:e215-e217.].

Early Onset Neovascular Inflammatory Vitreoretinopathy Due to a De Novo CAPN5 Mutation: Report of a Case.

Autosomal dominant neovascular inflammatory vitreoretinopathy (ADNIV) is a rare autoimmune condition that typically presents as progressive uveitis and vitreoretinal degeneration between the second and third decades of life. Though traditionally attributed to inherited mutations of the CAPN5 gene, few reports of de novo variants exist. This report of vision and hearing loss in a 3 year-old girl describes the youngest documented case of ADNIV due to a de novo pathogenic c.865C>T (p.Arg289Trp) CAPN5 variant, illustrating the early stages of this enigmatic disease process.

Use of percutaneous bleomycin sclerotherapy for orbital lymphatic malformations.

PURPOSE: Bleomycin sclerotherapy has been shown to be a viable treatment for lymphatic malformations. However, its use for these lesions confined to the orbit is becoming increasingly documented in the literature. In this study, we summarize the clinical manifestations and outcomes observed following percutaneous bleomycin sclerotherapy for orbital lymphatic malformation. METHODS: A 5-year retrospective chart review of patients with clinical, radiographic, and/or biopsy-confirmed diagnoses of orbital lymphatic malformation that received bleomycin sclerotherapy was conducted at the Emory Hospital and Clinics. Data examined included patient demographics, patient history and symptoms, clinical findings, radiographic findings, route of bleomycin delivery, and outcome. RESULTS: Of the 10 patients who met inclusion criteria, the median age of treatment was 7 years. The most common presenting symptoms included vision change and proptosis. Nine of 10 patients demonstrated macrocysts (>1 cm) on imaging. Seven of 10 patients had histories of prior interventions including resections, cyst drainage, and debulking. Because 2 of these 10 patients were lost to follow-up, 8 patients remained for post-procedural evaluation. Four of these eight showed improvement of visual acuity after post-bleomycin sclerotherapy. In seven of eight patients, extraocular motility either improved or remained stable. Pretreatment and posttreatment exophthalmometer measurements obtained in four patients revealed an average improvement in proptosis of 65% from their average pretreatment measurements. CONCLUSIONS: Our findings suggest that percutaneous bleomycin sclerotherapy is a viable option for treatment of orbital lymphatic malformations, with potentially greater benefit to those with macrocystic features.

TrkB signalling pathway mediates the protective effects of exercise in the diabetic rat retina.

Diabetic retinopathy is a leading cause of vision loss. Treatment options for early retinopathy are sparse. Exercise protects dying photoreceptors in models of retinal degeneration, thereby preserving vision. We tested the protective effects of exercise on retinal and cognitive deficits in a type 1 diabetes model and determined whether the TrkB pathway mediates this effect. Hyperglycaemia was induced in Long Evans rats via streptozotocin injection (STZ; 100 mg/kg). Following confirmed hyperglycaemia, both control and diabetic rats underwent treadmill exercise for 30 min, 5 days/week at 0 m/min (inactive groups) or 15 m/min (active groups) for 8 weeks. A TrkB receptor antagonist (ANA-12), or vehicle, was injected 2.5 h before exercise training. We measured spatial frequency and contrast sensitivity using optokinetic tracking biweekly post-STZ; retinal function using electroretinography at 4 and 8 weeks; and cognitive function and exploratory behaviour using Y-maze at 8 weeks. Retinal neurotrophin-4 was measured using ELISA. Compared with non-diabetic controls, diabetic rats showed significantly reduced spatial frequency and contrast sensitivity, delayed electroretinogram oscillatory potential and flicker implicit times and reduced cognitive function and exploratory behaviour. Exercise interventions significantly delayed the appearance of all deficits, except for exploratory behaviour. Treatment with ANA-12 significantly reduced this protection, suggesting a TrkB-mediated mechanism. Despite this, no changes in retinal neurotrohin-4 were observed with diabetes or exercise. Exercise protected against early visual and cognitive dysfunction in diabetic rats, suggesting that exercise interventions started after hyperglycaemia diagnosis may be a beneficial treatment. The translational potential is high, given that exercise treatment is non-invasive, patient controlled and inexpensive.