Konstantina Sampani MD , Mohamed Ashraf MD, PhD , Cloyd M. Pitoc BS , Jae Rhee BS , Ann M. Tolson BS , Jerry D. Cavallerano OD, PhD , Jennifer K. Sun MD, MPH , Lloyd Paul Aiello MD, PhD , Paolo S. Silva MD
{"title":"Association of Retinal Oximetry with Peripheral Diabetic Retinopathy Lesions and Nonperfusion on Ultra-widefield Angiography","authors":"Konstantina Sampani MD , Mohamed Ashraf MD, PhD , Cloyd M. Pitoc BS , Jae Rhee BS , Ann M. Tolson BS , Jerry D. Cavallerano OD, PhD , Jennifer K. Sun MD, MPH , Lloyd Paul Aiello MD, PhD , Paolo S. Silva MD","doi":"10.1016/j.xops.2024.100686","DOIUrl":"10.1016/j.xops.2024.100686","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the association of retinal ischemia measured using retinal oximetry with retinal nonperfusion and predominantly peripheral lesions on ultra-widefield (UWF) fluorescein angiography (FA PPL).</div></div><div><h3>Design</h3><div>Prospective single-center, image evaluation study.</div></div><div><h3>Participants</h3><div>Images from 42 eyes from 21 participants with diabetes.</div></div><div><h3>Methods</h3><div>Ultra-widefield images were evaluated to determine diabetic retinopathy (DR) severity. Ultra-widefield FA images were used to measure nonperfusion area (NPA, mm<sup>2</sup>) and nonperfusion index (NPI) and FA PPL presence. Retinal oximetry was performed to measure venous oxygen saturation (VO<sub>2</sub>, %) and arteriovenous difference (A-V, %) within a 2-disc diameter ring centered on the optic disc.</div></div><div><h3>Main Outcome Measures</h3><div>Nonperfusion area, NPI, and presence of FA PPL.</div></div><div><h3>Results</h3><div>Mean age was 40.7 ± 10.4 years, diabetes duration 21.4 ± 10.0 years, hemoglobin A1c (HbA1c) 7.7 ± 1.0, 33.3% (14) were female, and 76.2% (32) had type 1. Distribution of DR on UWF color imaging was no-DR 9.5% (4); mild 45.2% (19), moderate 21.4% (9), and severe 9.5% (4) nonproliferative DR; and proliferative DR 14.3 (6) with FA PPL present in 25 (59.5%). Mean NPA/NPI was associated with increasing DR severity (<em>P</em> = 0.0014/0.0018), even after correction for diabetes duration and HbA1c (<em>P</em> = 0.0029/0.0025). In multivariate analysis adjusting for diabetes duration, HbA1c, and DR severity, the presence of FA PPL was associated with increasing VO<sub>2</sub> and decreasing A-V (VO<sub>2</sub>; <em>P</em> = 0.03, A-V; <em>P</em> = 0.009).</div></div><div><h3>Conclusions</h3><div>Past studies have established an increased risk of DR progression with the presence of FA PPL. These data show that FA PPL presence is associated with retinal oximetry measures consistent with the presence of venous shunting or reduced retinal oxygen consumption, possibly indicative of greater areas of retinal ischemia. These findings highlight the value of retinal oximetry as a noninvasive measure of retinal ischemia and as a potential marker for increased risk of DR worsening.</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 100686"},"PeriodicalIF":3.2,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143619090","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}
Jeong Hyun Lee MD, MSc , Joo Young Shin MD, PhD , Martha Kim MD, PhD , Kyoung Min Lee MD, PhD , Sohee Oh PhD , Seok Hwan Kim MD, PhD , Ho-Kyung Choung MD, PhD , Jeeyun Ahn MD, PhD
{"title":"Changes in Choroidal Thickness of Healthy Children during Myopia Progression over 4 Years: Boramae Myopia Cohort Study Report 5","authors":"Jeong Hyun Lee MD, MSc , Joo Young Shin MD, PhD , Martha Kim MD, PhD , Kyoung Min Lee MD, PhD , Sohee Oh PhD , Seok Hwan Kim MD, PhD , Ho-Kyung Choung MD, PhD , Jeeyun Ahn MD, PhD","doi":"10.1016/j.xops.2024.100688","DOIUrl":"10.1016/j.xops.2024.100688","url":null,"abstract":"<div><h3>Objective</h3><div>To investigate the relationship between choroidal thickness and myopia by analyzing change in choroidal thickness over time in children with myopia progression.</div></div><div><h3>Design</h3><div>Retrospective cohort study.</div></div><div><h3>Participants</h3><div>Children with myopia.</div></div><div><h3>Methods</h3><div>Spherical equivalent (SE), axial length, and choroidal thickness were measured every 2 years during course of 4 years in children with myopia enrolled in a prospective cohort study. Choroidal thickness was evaluated at 13 points on the ETDRS grid, and its longitudinal changes as myopia progresses were analyzed. Patients were categorized into 2 subgroups: progression group (SE change ≤ −0.5 diopters [D] over 2 years) and stable group (SE change > −0.5 D over 2 years).</div></div><div><h3>Main Outcome Measures</h3><div>Spherical equivalent of refractive errors, axial length, and choroidal thickness.</div></div><div><h3>Results</h3><div>A total of 46 eyes from 23 participants were included, with a mean baseline age of 9.6 ± 1.7 years. The SE values at baseline, 2-year follow-up, and 4-year follow-up were −4.26 ± 2.34 D, −5.62 ± 2.45 D, and −8.67 ± 2.47 D, respectively, indicating an average myopia progression of 4.41 D over the 4-year period. During the initial 2 years, no significant thinning of choroidal thickness was observed at any of the 13 measured points. However, during the following 2 years, significant choroidal thinning was identified at 9 of the 13 points (<em>P</em> < 0.05). In the subgroup analysis of the subsequent 2 years, the progression group exhibited significant thinning at 8 points, while the stable group still showed no significant changes in choroidal thickness at any point.</div></div><div><h3>Conclusions</h3><div>In the early phase of myopia progression within moderate degree, choroidal thickness remained unchanged. However, when progressed to high myopia, significant choroidal thinning occurred, specifically in the progression group. In contrast, the stable group maintained consistent choroidal thickness throughout the study. These results suggest that choroidal thinning in children varies according to the degree of myopia that develops.</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 100688"},"PeriodicalIF":3.2,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143510871","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}
Daniel Wang BA , Bonnie Sklar MD , James Tian MD , Rami Gabriel MD , Matthew Engelhard MD, PhD , Ryan P. McNabb PhD , Anthony N. Kuo MD
{"title":"Improving Artificial Intelligence–based Microbial Keratitis Screening Tools Constrained by Limited Data Using Synthetic Generation of Slit-Lamp Photos","authors":"Daniel Wang BA , Bonnie Sklar MD , James Tian MD , Rami Gabriel MD , Matthew Engelhard MD, PhD , Ryan P. McNabb PhD , Anthony N. Kuo MD","doi":"10.1016/j.xops.2024.100676","DOIUrl":"10.1016/j.xops.2024.100676","url":null,"abstract":"<div><h3>Objective</h3><div>We developed a novel slit-lamp photography (SLP) generative adversarial network (GAN) model using limited data to supplement and improve the performance of an artificial intelligence (AI)–based microbial keratitis (MK) screening model.</div></div><div><h3>Design</h3><div>Cross-sectional study.</div></div><div><h3>Subjects</h3><div>Slit-lamp photographs of 67 healthy and 36 MK eyes were prospectively and retrospectively collected at a tertiary care ophthalmology clinic at a large academic institution.</div></div><div><h3>Methods</h3><div>We trained the GAN model StyleGAN2-ADA on healthy and MK SLPs to generate synthetic images. To assess synthetic image quality, we performed a visual Turing test. Three cornea fellows tested their ability to identify 20 images each of (1) real healthy, (2) real diseased, (3) synthetic healthy, and (4) synthetic diseased. We also used Kernel Inception Distance (KID) to quantitatively measure realism and variation of synthetic images. Using the same dataset used to train the GAN model, we trained 2 DenseNet121 AI models to grade SLP images as healthy or MK with (1) only real images and (2) real supplemented with GAN-generated images.</div></div><div><h3>Main Outcome Measures</h3><div>Classification performance of MK screening models trained with only real images compared to a model trained with both limited real and supplemented synthetic GAN images.</div></div><div><h3>Results</h3><div>For the visual Turing test, the fellows on average rated synthetic images as good quality (83.3% ± 12.0% of images), and synthetic and real images were found to depict pertinent anatomy and pathology for accurate classification (96.3% ± 2.19% of images). These experts could distinguish between real and synthetic images (accuracy: 92.5% ± 9.01%). Analysis of KID score for synthetic images indicated realism and variation. The MK screening model trained on both limited real and supplemented synthetic data (area under the receiver–operator characteristic curve: 0.93, bootstrapping 95% CI: 0.77–1.0) outperformed the model trained with only real data (area under the receiver–operator characteristic curve: 0.76, 95% CI: 0.50–1.0), with an improvement of 0.17 (95% CI: 0–0.4; 2-tailed <em>t</em> test <em>P</em> = 0.076).</div></div><div><h3>Conclusions</h3><div>Artificial intelligence–based MK classification may be improved by supplementation of limited real training data with synthetic data generated by GANs.</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 100676"},"PeriodicalIF":3.2,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551928","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}
Edmund Tsui MD, MS , Nicholas J. Jackson PhD, MPH , Judy L. Chen MD , Gary N. Holland MD
{"title":"Establishment of a Standard Technique for Determining Laser Flare Photometry Values during Assessment of Intraocular Inflammation","authors":"Edmund Tsui MD, MS , Nicholas J. Jackson PhD, MPH , Judy L. Chen MD , Gary N. Holland MD","doi":"10.1016/j.xops.2024.100690","DOIUrl":"10.1016/j.xops.2024.100690","url":null,"abstract":"<div><h3>Purpose</h3><div>The purposes of this study were to determine whether 2 laser flare photometry (LFP) devices produce similar results using various statistical techniques for determining a single, final LFP value for anterior chamber protein level; to determine whether a traditional technique adequately addresses outlier measurements; and to identify a simpler technique that produces LFP values similar to the traditional technique.</div></div><div><h3>Design</h3><div>Prospective cross-sectional study.</div></div><div><h3>Participants</h3><div>Patients at a tertiary referral center who have histories of uveitis in 1 or both eyes.</div></div><div><h3>Methods</h3><div>Seven LFP measurements were obtained on 200 eyes (100 patients) using Kowa FM-500 and FM-700 laser flare photometers (Kowa Company, Ltd). For each device, the final LFP values were determined using 4 statistical outlier removal techniques on all measurements and by means and medians for all 7 and for the first 5 measurements. The final LFP values by each technique were compared between the devices. Using the FM-700, host- and disease-related factors (age, sex, laterality, uveitis history, uveitis activity, and LFP values) were evaluated for their effects on outliers. Laser flare photometry values by each technique were compared with the LFP value by the traditional technique. Intraclass correlation coefficient (ICC) was used to compare values.</div></div><div><h3>Main Outcome Measures</h3><div>Laser flare photometry values by various techniques.</div></div><div><h3>Results</h3><div>The mean age of participants was 49.5 ± 19.1 years; 72% were female. Final LFP values did not vary meaningfully between the 2 devices for any technique (all ICCs: 0.81). With 5 measurements, no subgroup factors influenced the presence of outliers, and no final LFP values varied meaningfully from the final LFP value determined by the traditional technique (ICC: 0.97–1.00).</div></div><div><h3>Conclusions</h3><div>The automated mean provided by the FM-700 device based on 5 consecutive measurements may be a suitable final LFP value for use in patient care and clinical research.</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 100690"},"PeriodicalIF":3.2,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551925","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}
David Mikhail MD(C), MSc(C) , Daniel Milad MD , Fares Antaki MD, CM , Karim Hammamji MD , Cynthia X. Qian MD , Flavio A. Rezende MD, PhD , Renaud Duval MD, FRCSC
{"title":"The Role of Artificial Intelligence in Epiretinal Membrane Care: A Scoping Review","authors":"David Mikhail MD(C), MSc(C) , Daniel Milad MD , Fares Antaki MD, CM , Karim Hammamji MD , Cynthia X. Qian MD , Flavio A. Rezende MD, PhD , Renaud Duval MD, FRCSC","doi":"10.1016/j.xops.2024.100689","DOIUrl":"10.1016/j.xops.2024.100689","url":null,"abstract":"<div><h3>Topic</h3><div>In ophthalmology, artificial intelligence (AI) demonstrates potential in using ophthalmic imaging across diverse diseases, often matching ophthalmologists' performance. However, the range of machine learning models for epiretinal membrane (ERM) management, which differ in methodology, application, and performance, remains largely unsynthesized.</div></div><div><h3>Clinical Relevance</h3><div>Epiretinal membrane management relies on clinical evaluation and imaging, with surgical intervention considered in cases of significant impairment. AI analysis of ophthalmic images and clinical features could enhance ERM detection, characterization, and prognostication, potentially improving clinical decision-making. This scoping review aims to evaluate the methodologies, applications, and reported performance of AI models in ERM diagnosis, characterization, and prognostication.</div></div><div><h3>Methods</h3><div>A comprehensive literature search was conducted across 5 electronic databases including Ovid MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Web of Science Core Collection from inception to November 14, 2024. Studies pertaining to AI algorithms in the context of ERM were included. The primary outcomes measured will be the reported design, application in ERM management, and performance of each AI model.</div></div><div><h3>Results</h3><div>Three hundred ninety articles were retrieved, with 33 studies meeting inclusion criteria. There were 30 studies (91%) reporting their training and validation methods. Altogether, 61 distinct AI models were included. OCT scans and fundus photographs were used in 26 (79%) and 7 (21%) papers, respectively. Supervised learning and both supervised and unsupervised learning were used in 32 (97%) and 1 (3%) studies, respectively. Twenty-seven studies (82%) developed or adapted AI models using images, whereas 5 (15%) had models using both images and clinical features, and 1 (3%) used preoperative and postoperative clinical features without ophthalmic images. Study objectives were categorized into 3 stages of ERM care. Twenty-three studies (70%) implemented AI for diagnosis (stage 1), 1 (3%) identified ERM characteristics (stage 2), and 6 (18%) predicted vision impairment after diagnosis or postoperative vision outcomes (stage 3). No articles studied treatment planning. Three studies (9%) used AI in stages 1 and 2. Of the 16 studies comparing AI performance to human graders (i.e., retinal specialists, general ophthalmologists, and trainees), 10 (63%) reported equivalent or higher performance.</div></div><div><h3>Conclusion</h3><div>Artificial intelligence–driven assessments of ophthalmic images and clinical features demonstrated high performance in detecting ERM, identifying its morphological properties, and predicting visual outcomes following ERM surgery. Future research might consider the validation of algorithms for clinical applicati","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 4","pages":"Article 100689"},"PeriodicalIF":3.2,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681442","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}
Cameron D. Bruner BS, MD , Ashraf M. Mahmoud BS , Cynthia J. Roberts PhD
{"title":"Eccentric Pathology in Keratoconus Exhibits Stiffer Biomechanical Response than Central Pathology","authors":"Cameron D. Bruner BS, MD , Ashraf M. Mahmoud BS , Cynthia J. Roberts PhD","doi":"10.1016/j.xops.2024.100682","DOIUrl":"10.1016/j.xops.2024.100682","url":null,"abstract":"<div><h3>Purpose</h3><div>To investigate the difference in biomechanical response metrics between central and eccentric pathology and compare axial vs. tangential curvature, as well as zonal vs. single-point values.</div></div><div><h3>Design</h3><div>Prospective, observational, cross-sectional study.</div></div><div><h3>Participants</h3><div>The study included 67 eyes of 41 subjects diagnosed with keratoconus (KCN).</div></div><div><h3>Methods</h3><div>Pentacam tomography and Corvis ST examinations were acquired, and disease severity was defined by maximum curvature, comparing single point of maximum anterior axial curvature (Kmax) vs. magnitude of surrounding 2 mm zonal value (ZKmax) on axial maps, vs. magnitude of steepest 2 mm zone on axial (CSpot_Axi) and tangential (CSpot_Tan) maps located by Cone Location and Magnitude Index (CLMI). Distance between the corneal center and Kmax (Kmax_dist) was compared to radial distance with CLMI (CRad_Axi and CRad_Tan). Single-point Kmax, ZKmax, and CLMI-derived zones were compared with biomechanical metrics via regression analyses, including stiffness parameter at first applanation (SP-A1), deformation amplitude ratio at 2 mm (DA Ratio), integrated inverse radius (IIR), and stress–strain index (SSI). Measurements were analyzed using paired <em>t</em> tests, with <em>t</em> tests between central and eccentric disease, and a significance threshold, <em>P</em> < 0.05.</div></div><div><h3>Main Outcome Measures</h3><div>Maximum curvature using axial vs. tangential curvature, zonal vs. single-point curvature, and corneal stiffness metrics compared with cone location.</div></div><div><h3>Results</h3><div>Significantly greater central pathology was found using tangential (58 central and 9 eccentric) vs. axial curvature (28 central and 39 eccentric). ZKmax was significantly different than CSpot_Axi and CSpot_Tan (<em>P</em> < 0.0001). CRad_Axi (1.53 ± 0.41 mm) was significantly greater (<em>P</em> < 0.001) than Kmax_dist (1.33 ± 0.56 mm) and CRad_Tan (0.99 ± 0.34 mm). Kmax (56.09 ± 8.99 diopter [D]) was significantly greater than ZKmax (51.81 ± 7.50 D). Regressions for ZKmax, CSpot_Axi, and CSpot_Tan were significantly negative to SP-A1, stiffness parameter at highest concavity, and SSI, whereas significantly positive to DA Ratio and IIR. Regressions for Kmax_dist, CRad_Axi, and CRad_Tan had significantly positive relationships to SSI and significantly negative relationships to DA Ratio and IIR.</div></div><div><h3>Conclusions</h3><div>Central pathology has greater frequency with tangential than axial curvature. Corneal stiffness increases as the distance of the cone from the center increases, consistent with the focal nature of KCN. Central stiffness decreases as cone curvature (disease severity) increases. Recommendation is to use zonal values with tangential curvature to evaluate the location of the greatest curvature and changes in curvature over time.</div></div><div><h3>Financial Disclosure(s)</h3>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100682"},"PeriodicalIF":3.2,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143611683","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}
Isdin Oke MD, MPH , Tobias Elze PhD , Joan W. Miller MD , Alice C. Lorch MD, MPH , Ankoor S. Shah MD, PhD , David G. Hunter MD, PhD
{"title":"Trends in Billing Secure Messages at Ophthalmology Practices across the United States","authors":"Isdin Oke MD, MPH , Tobias Elze PhD , Joan W. Miller MD , Alice C. Lorch MD, MPH , Ankoor S. Shah MD, PhD , David G. Hunter MD, PhD","doi":"10.1016/j.xops.2024.100683","DOIUrl":"10.1016/j.xops.2024.100683","url":null,"abstract":"","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100683"},"PeriodicalIF":3.2,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551761","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}
Mohd. Afzal Khan BSc, BDS , Gehan Fatima BDS, PhD , Akm Ashiquzzaman MS , Sang Seong Kim PhD , Hyuksang Kwon PhD , Young Ro Kim PhD , Euiheon Chung PhD
{"title":"Evaluating the Preclinical Efficacy of Photobiomodulation in Alleviating Neuropathic Corneal Pain: A Behavioral Study","authors":"Mohd. Afzal Khan BSc, BDS , Gehan Fatima BDS, PhD , Akm Ashiquzzaman MS , Sang Seong Kim PhD , Hyuksang Kwon PhD , Young Ro Kim PhD , Euiheon Chung PhD","doi":"10.1016/j.xops.2024.100680","DOIUrl":"10.1016/j.xops.2024.100680","url":null,"abstract":"<div><h3>Purpose</h3><div>Neuropathic corneal pain (NCP) is a debilitating condition characterized by persistent pain due to corneal nerve damage or dysfunction. Millions of individuals and their families endure the significant impact of chronic pain. Effective management strategies are crucial yet limited, prompting the exploration of innovative treatments such as photobiomodulation (PBM).</div></div><div><h3>Design</h3><div>In vivo preclinical therapeutics investigation in mice.</div></div><div><h3>Subjects</h3><div>Thy1-YFP mice.</div></div><div><h3>Methods</h3><div>This study evaluates the efficacy of PBM in treating NCP across 4 animal models: normal control, sham control, pulled nerve, and full transection (FT). Behavioral assessments, including the von Frey test (VFT) for mechanical sensitivity and the eye-wiping test (EWT) for chemical sensitivity, were employed to evaluate the therapeutic impact of PBM till day 56 (D-1, D1, D3, D5, D7, D14, D28, D42, and D56).</div></div><div><h3>Main Outcome Measures</h3><div>Advances in therapeutic approach for NCP through the potential of PBM.</div></div><div><h3>Results</h3><div>Photobiomodulation significantly reduced behavioral manifestations of pain in the pulled nerve model (VFT: no PBM [D1 = 0.043 ± 0.044, D56 = 0.05 ± 0.014] and PBM [D1 = 0.050 ± 0.008 {<em>P</em> value = 0.18}, D56 = 0.09 ± 0.014 {<em>P</em> value = 0.02}], EWT: no PBM [D1 = 11.96 ± 0.47, D56 = 12.11 ± 0.15] and PBM [D1 = 11.73 ± 0.18 {<em>P</em> value = 0.2}, D56 = 11.22 ± 0.31] [<em>P</em> value = 0.01]) and FT model (VFT: no PBM [D1 = 0.022 ± 0.0028, D56 = 0.023 ± 0.0047] and PBM [D1 = 0.024 ± 0.0028 {<em>P</em> value = 0.2}, D56 = 0.073 ± 0.0094] [<em>P</em> value = 0.02]), EWT: no PBM [D1 = 13.1 ± 0.14, D56 = 13.36 ± 0.30] and PBM [D1 = 12.86 ± 0.41, {<em>P</em> value = 0.2}, D56 = 12.53 ± 0.41] [<em>P</em> value = 0.04]}, suggesting an effective reduction of pain sensitivity and an increase in corneal nerve function. The temporal patterns also suggest that early intervention with PBM, initiated shortly after nerve injury, may be crucial for preventing the chronic progression of NCP.</div></div><div><h3>Conclusions</h3><div>These outcomes support PBM as a promising nonpharmacologic intervention for NCP; this not only reinforces the potential of PBM in NCP treatment but also provides a foundation for future clinical applications in managing corneal 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 100680"},"PeriodicalIF":3.2,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551933","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}
Akshay Prashant Agnihotri MS, DNB , Ines Doris Nagel MD , Jose Carlo M. Artiaga MD, FICO , Ma. Carmela B. Guevarra MD , George Michael N. Sosuan MD , Fritz Gerald P. Kalaw MD
{"title":"Large Language Models in Ophthalmology: A Review of Publications from Top Ophthalmology Journals","authors":"Akshay Prashant Agnihotri MS, DNB , Ines Doris Nagel MD , Jose Carlo M. Artiaga MD, FICO , Ma. Carmela B. Guevarra MD , George Michael N. Sosuan MD , Fritz Gerald P. Kalaw MD","doi":"10.1016/j.xops.2024.100681","DOIUrl":"10.1016/j.xops.2024.100681","url":null,"abstract":"<div><h3>Purpose</h3><div>To review and evaluate the current literature on the application and impact of large language models (LLMs) in the field of ophthalmology, focusing on studies published in high-ranking ophthalmology journals.</div></div><div><h3>Design</h3><div>This is a retrospective review of published articles.</div></div><div><h3>Participants</h3><div>This study did not involve human participation.</div></div><div><h3>Methods</h3><div>Articles published in the first quartile (Q1) of ophthalmology journals on Scimago Journal & Country Rank discussing different LLMs up to June 7, 2024, were reviewed, parsed, and analyzed.</div></div><div><h3>Main Outcome Measures</h3><div>All available articles were parsed and analyzed, which included the article and author characteristics and data regarding the LLM used and its applications, focusing on its use in medical education, clinical assistance, research, and patient education.</div></div><div><h3>Results</h3><div>There were 35 Q1-ranked journals identified, 19 of which contained articles discussing LLMs, with 101 articles eligible for review. One-third were original investigations (32%; 32/101), with an average of 5.3 authors per article. The United States (50.4%; 51/101) was the most represented country, followed by the United Kingdom (25.7%; 26/101) and Canada (16.8%; 17/101). ChatGPT was the most used LLM among the studies, with different versions discussed and compared. Large language model applications were discussed relevant to their implications in medical education, clinical assistance, research, and patient education.</div></div><div><h3>Conclusions</h3><div>The numerous publications on the use of LLM in ophthalmology can provide valuable insights for stakeholders and consumers of these applications. Large language models present significant opportunities for advancement in ophthalmology, particularly in team science, education, clinical assistance, and research. Although LLMs show promise, they also show challenges such as performance inconsistencies, bias, and ethical concerns. The study emphasizes the need for ongoing artificial intelligence improvement, ethical guidelines, and multidisciplinary collaboration.</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 100681"},"PeriodicalIF":3.2,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551760","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}
Rohith Ravindranath MS, Joel Naor MD, MS, Sophia Y. Wang MD, MS
{"title":"Artificial Intelligence Models to Identify Patients at High Risk for Glaucoma Using Self-reported Health Data in a United States National Cohort","authors":"Rohith Ravindranath MS, Joel Naor MD, MS, Sophia Y. Wang MD, MS","doi":"10.1016/j.xops.2024.100685","DOIUrl":"10.1016/j.xops.2024.100685","url":null,"abstract":"<div><h3>Purpose:</h3><div>Early glaucoma detection is key to preventing vision loss, but screening often requires specialized eye examination or photography, limiting large-scale implementation. This study sought to develop artificial intelligence models that use self-reported health data from surveys to prescreen patients at high risk for glaucoma who are most in need of glaucoma screening with ophthalmic examination and imaging.</div></div><div><h3>Design:</h3><div>Cohort study.</div></div><div><h3>Participants:</h3><div>Participants enrolled from May 1, 2018, to July 1, 2022, in the nationwide All of Us Research Program who were ≥18 years of age, had ≥2 eye-related diagnoses in their electronic health record (EHR), and submitted surveys with self-reported health history.</div></div><div><h3>Methods:</h3><div>We developed models to predict the risk of glaucoma, as determined by EHR diagnosis codes, using 3 machine learning approaches: (1) penalized logistic regression, (2) XGBoost, and (3) a fully connected neural network. Glaucoma diagnosis was identified based on International Classification of Diseases codes extracted from EHR data. An 80/20 train–test split was implemented, with cross-validation employed for hyperparameter tuning. Input features included self-reported demographics, general health, lifestyle factors, and family and personal medical history.</div></div><div><h3>Main Outcome Measures:</h3><div>Models were evaluated using standard classification metrics, including area under the receiver operating characteristic curve (AUROC).</div></div><div><h3>Results:</h3><div>Among the 8205 patients, 873 (10.64%) were diagnosed with glaucoma. Across models, AUROC scores for identifying which patients had glaucoma from survey health data ranged from 0.710 to 0.890. XGBoost achieved the highest AUROC of 0.890 (95% confidence interval [CI]: 0.860–0.910). Logistic regression followed with an AUROC of 0.772 (95% CI: 0.753–0.795). Explainability studies revealed that key features included traditionally recognized risk factors for glaucoma, such as age, type 2 diabetes, and a family history of glaucoma.</div></div><div><h3>Conclusions:</h3><div>Machine and deep learning models successfully utilized health data from self-reported surveys to predict glaucoma diagnosis without additional data from ophthalmic imaging or eye examination. These models may eventually enable prescreening for glaucoma in a wide variety of low-resource settings, after which high-risk patients can be referred for targeted screening using more specialized ophthalmic examination or imaging.</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 100685"},"PeriodicalIF":3.2,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592110","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}