Ophthalmology science最新文献

{"title":"Bio-interventional Uveoscleral Outflow Enhancement Surgery for Primary Open-Angle Glaucoma: 2-Year Results of Cyclodialysis with Scleral Allograft Reinforcement","authors":"Ernesto Calvo MD ,&nbsp;Ticiana De Francesco MD ,&nbsp;Lautaro Vera MD ,&nbsp;Farrell Tyson MD, FACS ,&nbsp;Robert N. Weinreb MD","doi":"10.1016/j.xops.2025.100727","DOIUrl":"10.1016/j.xops.2025.100727","url":null,"abstract":"<div><h3>Objective</h3><div>To evaluate safety and efficacy of the scleral allograft–reinforced cyclodialysis through 24 months of follow-up.</div></div><div><h3>Design</h3><div>Interventional single-center case series.</div></div><div><h3>Participants</h3><div>Thirty-one eyes with primary open-angle glaucoma and visually significant cataracts underwent bio-interventional cyclodialysis surgery with scleral allograft reinforcement combined with phacoemulsification.</div></div><div><h3>Intervention</h3><div>Uveoscleral outflow enhancement surgery comprised of cyclodialysis with sequential bio-reinforcement with a scleral allograft combined with phacoemulsification.</div></div><div><h3>Main Outcome Measures</h3><div>The primary outcome was the proportion of eyes achieving ≥20% intraocular pressure (IOP) reduction with same or fewer medications compared with baseline. Secondary outcomes included the mean change in medicated IOP and mean number of IOP-lowering medications compared with baseline. Adverse events were also recorded and evaluated throughout the study.</div></div><div><h3>Results</h3><div>The primary outcome was achieved in 74% of the eyes, and there was a mean IOP reduction of 34% compared with baseline. Baseline mean medicated IOP was 21.9 ± 4.92 mmHg on 1.22 ± 1.29 IOP-lowering medications. At 12 months postoperation, mean IOP was 12.62 ± 2.63 on 0.55 ± 0.52 glaucoma medications. The procedure was well tolerated, and there were no serious ocular adverse events.</div></div><div><h3>Conclusions</h3><div>Uveoscleral outflow enhancement can be successfully achieved at the time of cataract surgery through bio-interventional cyclodialysis and scleral allograft reinforcement to lower IOP in patients with primary open-angle glaucoma.</div></div><div><h3>Financial Disclosure(s)</h3><div>The author(s) have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100727"},"PeriodicalIF":3.2,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Role of Menopause and Its Association with the Apolipoprotein E4 Allele for Age at Diagnosis of Glaucoma in Women","authors":"Yan Shi MD, PhD ,&nbsp;William Liu ,&nbsp;Junming Hu PhD ,&nbsp;Wei Qiao Qiu MD, PhD ,&nbsp;Xinyue He MD, PhD ,&nbsp;Yan Gao MD ,&nbsp;Xiaoling Zhang MD, PhD ,&nbsp;Zhigang Fan MD, PhD","doi":"10.1016/j.xops.2025.100726","DOIUrl":"10.1016/j.xops.2025.100726","url":null,"abstract":"<div><h3>Objective</h3><div>To explore the impact of menopause for age at diagnosis (AAD) of glaucoma in women and illustrate its interaction with the apolipoprotein E (<em>APOE</em>) E4 allele.</div></div><div><h3>Design</h3><div>A retrospective, case-only analysis using the UK Biobank participants with complete data (2006–2010) for analysis.</div></div><div><h3>Participants</h3><div>One thousand three hundred fifty-eight female glaucoma patients.</div></div><div><h3>Methods</h3><div>Multivariable-adjusted associations of AAD of glaucoma, <em>APOE</em> E4 allele, age of menopause, and hormone replacement therapy (HRT) were analyzed by linear mixed model (LMM) analyses across groups stratified by whether glaucoma developed before or after menopause and whether or not HRT was used.</div></div><div><h3>Main Outcome Measures</h3><div>Age at diagnosis of glaucoma, age of menopause, <em>APOE</em> E4 allele, and HRT information.</div></div><div><h3>Results</h3><div>The age-adjusted univariate LMM showed that later menopause was significantly associated with an older AAD of glaucoma in both the overall cohort and subgroups where glaucoma developed before or after menopause (model 1, all <em>P</em> &lt; 0.05). The age-adjusted multivariate LMM found that carrying the <em>APOE</em> E4 allele combined with later menopause significantly increased the AAD of glaucoma in patients diagnosed before menopause (model 3: β_{age of menopause} = 0.711 ± 0.074, <em>P</em> &lt; 0.001; β_e4 = 1.406 ± 0.596, <em>P</em> = 0.019; model 1 vs. model 3: <em>P</em> = 0.018). No similar association was observed in patients diagnosed after menopause (<em>P</em> &gt; 0.05). Additionally, the age-adjusted univariate LMM showed that HRT was associated with an older AAD of glaucoma (model 4: β_HRT = 1.239 ± 0.368, <em>P</em> = 0.001), with this effect being more pronounced in patients with later menopause (model 5: β_HRT = 1.625 ± 0.356, <em>P</em> &lt; 0.001; β_{age of menopause} = 0.301 ± 0.033, <em>P</em> &lt; 0.001; model 4 vs. model 5: <em>P</em> &lt; 0.001).</div></div><div><h3>Conclusions</h3><div>Later menopause was associated with an older AAD of glaucoma, with the <em>APOE</em> E4 allele providing increased protection against glaucoma in those diagnosed before, but not after, menopause. The protective effect of later menopause was also enhanced by HRT use after menopause. These findings underscore the interaction of hormonal status and <em>APOE</em> genotype in glaucoma onset, potentially guiding the prevention or management of glaucoma and other age-related health conditions in women.</div></div><div><h3>Financial Disclosure(s)</h3><div>The author(s) have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100726"},"PeriodicalIF":3.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Autonomous Screening for Diabetic Macular Edema Using Deep Learning Processing of Retinal Images","authors":"Idan Bressler ,&nbsp;Rachelle Aviv ,&nbsp;Danny Margalit ,&nbsp;Gal Yaakov Cohen MD ,&nbsp;Tsontcho Ianchulev MD, MPH ,&nbsp;Shravan V. Savant MD ,&nbsp;David J. Ramsey MD, PhD ,&nbsp;Zack Dvey-Aharon PhD","doi":"10.1016/j.xops.2025.100722","DOIUrl":"10.1016/j.xops.2025.100722","url":null,"abstract":"<div><h3>Objective</h3><div>To develop and validate a deep learning model for diabetic macular edema (DME) detection using color fundus imaging, which is applicable in a diverse, multidevice clinical setting.</div></div><div><h3>Design</h3><div>Evaluation of diagnostic test or technology.</div></div><div><h3>Subjects</h3><div>A deep learning model was trained for DME detection using the EyePACS dataset, consisting of 32 049 images from 15 892 patients. The average age was 55.02%, and 51% of the patients were women.</div></div><div><h3>Methods</h3><div>Data were randomly assigned, by participant, into development (n = 14 246) and validation (n = 1583) sets. Analysis was conducted on the single image, eye, and patient levels. Model performance was evaluated using sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC). Independent validation was further performed on the Indian Diabetic Retinopathy Image Dataset, as well as on new data.</div></div><div><h3>Main Outcome Measures</h3><div>Sensitivity, specificity, and AUC.</div></div><div><h3>Results</h3><div>At the image level, a sensitivity of 0.889 (95% confidence interval [CI]: 0.878, 0.900), a specificity of 0.889 (95% CI: 0.877, 0.900), and an AUC of 0.954 (95% CI: 0.949, 0.959) were achieved. At the eye level, a sensitivity of 0.905 (95% CI: 0.890, 0.920), a specificity of 0.902 (95% CI: 0.890, 0.913), and an AUC of 0.964 (95% CI: 0.958, 0.969) were achieved. At the patient level, a sensitivity of 0.900 (95% CI: 0.879, 0.917), a specificity of 0.900 (95% CI: 0.883, 0.911), and an AUC of 0.962 (95% CI: 0.955, 0.968) were achieved.</div></div><div><h3>Conclusions</h3><div>Diabetic macular edema can be detected from color fundus imaging with high performance on all analysis metrics. Automatic DME detection may simplify screening, leading to more encompassing screening for diabetic patients. Further prospective studies are necessary.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100722"},"PeriodicalIF":3.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Long-Term Monitoring of Corneal Grafts Via Anterior Segment OCT Pachymetry Maps","authors":"Anastasia Neokleous MD, MSc ,&nbsp;Neofytos Michail MD ,&nbsp;Fedonas Herodotou MD ,&nbsp;Aikaterini Athanasiadou MSc ,&nbsp;Stylianos Christodoulou MD, MSc ,&nbsp;Dimitris Kola MD ,&nbsp;Klea Panayidou PhD ,&nbsp;Georgina Hadjilouka MD, PhD ,&nbsp;Sotiria Palioura MD, PhD","doi":"10.1016/j.xops.2025.100724","DOIUrl":"10.1016/j.xops.2025.100724","url":null,"abstract":"<div><h3>Purpose</h3><div>To assess the efficacy of anterior segment OCT (AS-OCT) in the long-term monitoring of corneal grafts and its integration into a hybrid remote care model for early detection and management of graft rejection or failure.</div></div><div><h3>Design</h3><div>Prospective cohort study.</div></div><div><h3>Participants</h3><div>Seventy-four patients (93 eyes) who underwent corneal transplantation from October 2021 to December 2023, with a follow-up period of ≥6 months.</div></div><div><h3>Methods</h3><div>Serial AS-OCT pachymetry maps and cross-sectional scans were performed at fixed postoperative intervals, and the findings were correlated with clinical signs of graft rejection or failure on slit lamp examination for thickness changes &gt;50 μm. A hybrid remote AS-OCT screening protocol was initiated 1 week postoperatively.</div></div><div><h3>Main Outcome Measures</h3><div>Diagnostic accuracy of AS-OCT, measured by specificity and sensitivity, in detecting graft rejection or failure through changes in corneal and graft thickness.</div></div><div><h3>Results</h3><div>Anterior segment OCT demonstrated high diagnostic accuracy with a specificity of 97.6% and a sensitivity of 88.9% in detecting graft rejection or failure. The mean central corneal thickness increase in cases resulting in graft rejection or failure was 82.7 ± 21.5 μm, a thickness change that is not discernible by slit lamp examination alone. The utility of AS-OCT in a hybrid remote monitoring model was demonstrated through 3 detailed case studies, highlighting improved clinical workflow and patient convenience without compromising postoperative outcomes.</div></div><div><h3>Conclusions</h3><div>Serial AS-OCT imaging serves as a robust, objective, and quantitative tool for postoperative surveillance of corneal grafts, significantly benefiting patient outcomes by allowing timely interventions. Integration of AS-OCT into a hybrid remote screening protocol supports comprehensive monitoring, complementing direct clinical evaluations and optimizing postoperative care.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100724"},"PeriodicalIF":3.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Population-Based Normative Reference for Retinal Microvascular Atlas","authors":"Mayinuer Yusufu MTI ,&nbsp;Algis J. Vingrys PhD ,&nbsp;Xianwen Shang PhD ,&nbsp;Lei Zhang PhD ,&nbsp;Danli Shi PhD, MD ,&nbsp;Nathan Congdon PhD, MD ,&nbsp;Mingguang He PhD, MD","doi":"10.1016/j.xops.2025.100723","DOIUrl":"10.1016/j.xops.2025.100723","url":null,"abstract":"<div><h3>Objective</h3><div>To establish the normative range of a comprehensive set of retinal vascular measurements to better understand their value as biomarkers for assessing ocular and systemic health.</div></div><div><h3>Design</h3><div>Cross-sectional study.</div></div><div><h3>Participants</h3><div>The study included 10 151 healthy participants from the UK Biobank.</div></div><div><h3>Methods</h3><div>Retina-based Microvascular Health Assessment System software was used to extract retinal vascular measurements, including caliber, complexity, density, branching angle, and tortuosity, differentiating between arteries and veins and between the macula and retinal periphery. In addition, we explored relationships between those measurements and health metrics, including age, systolic blood pressure (SBP), body mass index, glycated hemoglobin, and intraocular pressure.</div></div><div><h3>Main Outcome Measures</h3><div>We reported the population normative range for 114 retinal vascular measurements, further stratified by sex and age.</div></div><div><h3>Results</h3><div>The mean values of central retinal artery equivalent and central retinal vein equivalent (CRVE) were 152 (standard deviation = 14.9) μm and 233 (21.5) μm, respectively. The mean value of fractal dimension (FD) was 1.77 (0.032), with arterial FD 1.53 (0.039) and venular FD 1.56 (0.025). Age and SBP showed the strongest associations with most retinal parameters among health metrics. Central retinal artery equivalent, CRVE, density, and complexity decreased with increasing age and SBP. Changes in arterial measurements with age and SBP were generally greater than those in venous measurements. Generalized additive models further revealed that observed associations were mainly linear.</div></div><div><h3>Conclusions</h3><div>By establishing population normative data for a comprehensive set of retinal vascular measurements, our study enables quantifiable approaches to better understand retinal vascular changes.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100723"},"PeriodicalIF":3.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Elevated Intraocular Pressure on Lamina Cribrosa Oxygenation: A Combined Experimental–Computational Study on Monkeys","authors":"Yuankai Lu PhD ,&nbsp;Yi Hua PhD ,&nbsp;Bingrui Wang PhD ,&nbsp;Qi Tian ,&nbsp;Fuqiang Zhong PhD ,&nbsp;Andrew Theophanous ,&nbsp;Shaharoz Tahir ,&nbsp;Po-Yi Lee PhD ,&nbsp;Ian A. Sigal PhD","doi":"10.1016/j.xops.2025.100725","DOIUrl":"10.1016/j.xops.2025.100725","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate how lamina cribrosa (LC) oxygenation is affected by tissue distortions resulting from elevated intraocular pressure (IOP).</div></div><div><h3>Design</h3><div>Experimental study on 4 monkeys, histology, and computational analysis.</div></div><div><h3>Subjects</h3><div>Four healthy monkey eyes with OCT scans at IOPs of 10 to 60 mmHg.</div></div><div><h3>Methods</h3><div>Intraocular pressure–induced LC tissue deformations of a healthy monkey were measured in vivo using OCT images and digital volume correlation analysis techniques. Three-dimensional eye-specific models of the LC vasculature of 4 healthy monkey eyes were reconstructed using histology. The models were then used to compute LC oxygenation, first as reconstructed (baseline), and then with the LC vessels distorted according to the OCT-derived deformations. Two biomechanics-based mapping techniques were evaluated: cross-sectional and isotropic. The hemodynamics and oxygenations of the 4 LC vessel networks were evaluated at IOPs up to 60 mmHg to quantify the effects of IOP on LC oxygen supply, assorting the extent of LC tissue mild and severe hypoxia.</div></div><div><h3>Main Outcome Measures</h3><div>Intraocular pressure–induced deformation, vasculature structure, blood supply, and LC oxygenation.</div></div><div><h3>Results</h3><div>Intraocular pressure–induced deformations reduced LC oxygenation significantly and substantially. More than 20% of LC tissue suffered from mild hypoxia when IOP reached 30 mmHg. Extreme IOP (&gt;50 mmHg) led to large severe hypoxia regions (&gt;30%) in the isotropic mapping cases.</div></div><div><h3>Conclusions</h3><div>Our calculations predicted that moderately elevated IOP can lead to mild hypoxia in a substantial part of the LC, which, if sustained chronically, may contribute to neural tissue damage. For extreme IOP elevations, severe hypoxia was predicted, which would likely cause more immediate damage. Our findings suggest that despite the remarkable LC vascular robustness, IOP-induced distortions can potentially contribute to glaucomatous neuropathy.</div></div><div><h3>Financial Disclosure(s)</h3><div>The author(s) have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100725"},"PeriodicalIF":3.2,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Code-Free Deep Learning Glaucoma Detection on Color Fundus Images","authors":"Daniel Milad MD ,&nbsp;Fares Antaki MDCM ,&nbsp;David Mikhail MD (C), MSc (C) ,&nbsp;Andrew Farah MDCM (C) ,&nbsp;Jonathan El-Khoury MD ,&nbsp;Samir Touma MD ,&nbsp;Georges M. Durr MD ,&nbsp;Taylor Nayman MD ,&nbsp;Clément Playout PhD (C) ,&nbsp;Pearse A. Keane MD, FRCOphth ,&nbsp;Renaud Duval MD","doi":"10.1016/j.xops.2025.100721","DOIUrl":"10.1016/j.xops.2025.100721","url":null,"abstract":"<div><h3>Objective</h3><div>Code-free deep learning (CFDL) allows clinicians with no coding experience to build their own artificial intelligence models. This study assesses the performance of CFDL in glaucoma detection from fundus images in comparison to expert-designed models.</div></div><div><h3>Design</h3><div>Deep learning model development, testing, and validation.</div></div><div><h3>Subjects</h3><div>A total of 101 442 labeled fundus images from the Rotterdam EyePACS Artificial Intelligence for Robust Glaucoma Screening (AIROGS) dataset were included.</div></div><div><h3>Methods</h3><div>Ophthalmology trainees without coding experience designed a CFDL binary model using the Rotterdam EyePACS AIROGS dataset of fundus images (101 442 labeled images) to differentiate glaucoma from normal optic nerves. We compared our results with bespoke models from the literature. We then proceeded to externally validate our model using 2 datasets, the Retinal Fundus Glaucoma Challenge (REFUGE) and the Glaucoma grading from Multi-Modality imAges (GAMMA) at 0.1, 0.3, and 0.5 confidence thresholds.</div></div><div><h3>Main Outcome Measures</h3><div>Area under the precision-recall curve (AuPRC), sensitivity at 95% specificity (SE@95SP), accuracy, area under the receiver operating curve (AUC), and positive predictive value (PPV).</div></div><div><h3>Results</h3><div>The CFDL model showed high performance metrics that were comparable to the bespoke deep learning models. Our single-label classification model had an AuPRC of 0.988, an SE@95SP of 95%, and an accuracy of 91% (compared with 85% SE@95SP for the top bespoke models). Using the REFUGE dataset for external validation, our model had an SE@95SP, AUC, PPV, and accuracy of 83%, 0.960%, 73% to 94%, and 95% to 98%, respectively, at the 0.1, 0.3, and 0.5 confidence threshold cutoffs. Using the GAMMA dataset for external validation at the same confidence threshold cutoffs, our model had an SE@95SP, AUC, PPV, and accuracy of 98%, 0.994%, 94% to 96%, and 94% to 97%, respectively.</div></div><div><h3>Conclusion</h3><div>The capacity of CFDL models to perform glaucoma screening using fundus images presents a compelling proof of concept, empowering clinicians to explore innovative model designs for broad glaucoma screening in the near future.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100721"},"PeriodicalIF":3.2,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Independent Evaluation of RETFound Foundation Model's Performance on Optic Nerve Analysis Using Fundus Photography","authors":"Maggie S. Chen ,&nbsp;Rohith Ravindranath MS ,&nbsp;Robert Chang MD ,&nbsp;Yukun Zhou PhD ,&nbsp;Pearse A. Keane MD FRCOphth ,&nbsp;Sophia Y. Wang MD, MS","doi":"10.1016/j.xops.2025.100720","DOIUrl":"10.1016/j.xops.2025.100720","url":null,"abstract":"<div><h3>Purpose</h3><div>This study evaluates RETFound, a retinal image foundation model, as a feature extractor for predicting optic nerve metrics like cup-to-disc ratio (CDR) and retinal nerve fiber layer (RNFL) thickness using an independent clinical dataset.</div></div><div><h3>Design</h3><div>Retrospective observational study.</div></div><div><h3>Participants</h3><div>Patients who underwent fundus photography and RNFL OCT at the Byers Eye Institute, Stanford University.</div></div><div><h3>Methods</h3><div>Fundus images were paired with RNFL OCT results where study dates were within 6 months of each other. Latent features from full-sized raw fundus images were extracted from RETFound and used as inputs for several linear regression models (Ridge, Lasso, Elastic Net, and ordinary least squares). Baseline models using pretrained VGG16 and Vision Transformers (ViTs) as feature extractors were also developed. All models were trained to perform single-output tasks (predicting CDR or average RNFL thickness) and multioutput tasks (predicting RNFL thickness at quadrants and clock hours). Data were split 80:20 at the patient level for training and validation.</div></div><div><h3>Main Outcome Measures</h3><div>Model predictions were evaluated on a test set using the metrics of <em>R</em>^<em>2</em>, mean absolute error, and root mean square error.</div></div><div><h3>Results</h3><div>Among the 463 unique participants, contributing 776 fundus–OCT data pairs, the mean age was 63 years (±18 years), with 57.24% being female (N = 265). RETFound models demonstrated strong performance on single-output tasks, achieving <em>R</em>^<em>2</em> values between 0.706 and 0.898 for CDR prediction and between 0.855 and 0.961 for average RNFL thickness prediction. Performance on multioutput tasks was less robust, with a highest <em>R</em>^<em>2</em> of 0.583 for clock-hour RNFL thickness prediction and an <em>R</em>^<em>2</em> of 0.811 for quadrant RNFL thickness prediction. RETFound models outperformed VGG16 and ViT models, which achieved maximum <em>R</em>^<em>2</em> of 0.731 and 0.687 in predicting RNFL thickness and CDR.</div></div><div><h3>Conclusions</h3><div>Machine learning models leveraging the massively pretrained RETFound foundation model could accurately predict CDR and average RNFL thickness from fundus photos on an independent clinical dataset. Although RETFound was not trained or fine-tuned for these optic nerve evaluation tasks, nevertheless, RETFound overcomes small dataset limitations and excels in specialized applications.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100720"},"PeriodicalIF":3.2,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sex Differences in Age-Related Changes in Retinal Arteriovenous Area Based on Deep Learning Segmentation Model","authors":"Mizuho Mitamura MD ,&nbsp;Michiyuki Saito MD, PhD ,&nbsp;Kanae Fukutsu MD, PhD ,&nbsp;Zhenyu Dong MD, PhD ,&nbsp;Ryo Ando MD, PhD ,&nbsp;Satoru Kase MD, PhD ,&nbsp;Kousuke Noda MD, PhD ,&nbsp;Ryosuke Shiba ,&nbsp;Naoki Isogai ,&nbsp;Mitsuru Dohke MD ,&nbsp;Manabu Kase MD, PhD ,&nbsp;Susumu Ishida MD, PhD","doi":"10.1016/j.xops.2025.100719","DOIUrl":"10.1016/j.xops.2025.100719","url":null,"abstract":"","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100719"},"PeriodicalIF":3.2,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of the Coronavirus Disease 2019 Pandemic on Initiation Therapy for Noninfectious Uveitis","authors":"Evan M. Chen MD,&nbsp;D. Claire Miller MS,&nbsp;Yuwei Sun MS,&nbsp;Anika Kumar BA,&nbsp;Jason Richards MPH,&nbsp;Nisha R. Acharya MD","doi":"10.1016/j.xops.2025.100718","DOIUrl":"10.1016/j.xops.2025.100718","url":null,"abstract":"<div><h3>Purpose</h3><div>Initial studies during the coronavirus disease 2019 (COVID-19) pandemic demonstrated a possible increased risk of COVID-19 infection and severe outcomes with prior or concurrent immunomodulatory therapy (IMT). The purpose of this study was to determine the impact of the COVID-19 pandemic on treatment patterns for noninfectious uveitis (NIU).</div></div><div><h3>Design</h3><div>Retrospective interrupted time series (ITS) analysis using Optum Labs Data Warehouse, a national deidentified health care database in the United States with administrative claims and electronic health record data.</div></div><div><h3>Participants</h3><div>Individuals with a new diagnosis of NIU from December 1, 2017, to December 31, 2020, with continuous enrollment ≥1 year before this diagnosis.</div></div><div><h3>Methods</h3><div>This study was divided into 3 time periods: prepandemic (December 1, 2017–November 30, 2019), early pandemic (March 1, 2020–December 31, 2020), and postvaccine period (January 1, 2021–September 30, 2021) corresponding to time before the pandemic, during the pandemic when no COVID-19 vaccine was available, and after widespread utilization of the vaccine began. Normalized prescription rates of uveitis therapies were modeled as an ITS. In the time-to-treatment analysis, Cox proportional hazard models were used to determine differences in likelihood of different modalities between time periods.</div></div><div><h3>Main Outcome Measures</h3><div>Temporal trends in the initial therapeutic choice for NIU.</div></div><div><h3>Results</h3><div>This study included 22 444 patients with a new NIU diagnosis. The average age was 61.9 (standard deviation 17.5) years, and 59.3% were female. There were no significant temporal breaks in prescribing trends for topical, local, and systemic corticosteroids or immunosuppressive therapy (disease-modifying antirheumatic drugs and biologics) between pandemic periods (all <em>P</em> &gt; 0.05) in ITS analysis. Overall, topical steroids were more likely to be prescribed in the early versus prepandemic period (hazard ratio [HR] 1.10; 95% confidence interval [CI] 1.06–1.15; <em>P</em> &lt; 0.001). Intraocular steroids also saw greater relative use during the early (HR 1.29; 95% CI 1.13–1.46; <em>P</em> &lt; 0.001) and postvaccine (HR 1.29; 95% CI 1.14–1.46; <em>P</em> &lt; 0.001) period. Use of IMTs increased in the postvaccine period compared with that in the prepandemic period (HR 1.25; 95% CI 1.07–1.46; <em>P</em> &lt; 0.001).</div></div><div><h3>Conclusions</h3><div>No significant differences in prescribing patterns for NIU were observed between pandemic periods. However, utilization of topical and local steroids for NIU was, overall, increased in the early compared with the prepandemic period.</div></div><div><h3>Financial Disclosure(s)</h3><div>The author(s) have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100718"},"PeriodicalIF":3.2,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}