AJO International最新文献

{"title":"What patients ask: Evaluating google search queries and webpage transparency in common ophthalmologic malignancies","authors":"Siddharth Gandhi ,&nbsp;Janmejay Buranpuri ,&nbsp;Michael Balas ,&nbsp;Rachel Curtis","doi":"10.1016/j.ajoint.2026.100224","DOIUrl":"10.1016/j.ajoint.2026.100224","url":null,"abstract":"<div><h3>Objective/Purpose</h3><div>To characterize the online information landscape for ocular malignancies by analyzing patient search queries and evaluating the transparency and readability of linked websites.</div></div><div><h3>Design</h3><div>Cross-sectional infodemiologic analysis. Searches were conducted on June 12, 2025, using a private, location-masked web browser.</div></div><div><h3>Subjects, Participants, and/or Controls</h3><div>A total of 300 Google “People Also Ask” (PAA) questions and 205 unique linked websites associated with three search terms: uveal melanoma, retinoblastoma, and ocular surface tumours (100 questions per term).</div></div><div><h3>Methods, Intervention, or Testing</h3><div>Questions were categorized by type and topic using an adapted Rothwell classification. Websites were classified by source (e.g., commercial, government) and evaluated using the JAMA Benchmark Criteria (transparency) and Flesch-Kincaid Reading Ease (FKRE) and Coleman-Liau Index (readability).</div></div><div><h3>Results</h3><div>Fact-based queries predominated (71 %). Search topics varied significantly by diagnosis: prognosis was the primary concern for uveal melanoma (17 %) and retinoblastoma (15 %), whereas technical details (24 %) dominated for ocular surface tumours. Commercial websites were the most visible source (44 %). Government sites achieved the highest transparency (mean JAMA Benchmark 3.30) yet were amongst the most difficult to read (FKRE 37.8). Conversely, commercial sites were more readable (FKRE 51.8) but less transparent (JAMA Benchmark 2.67). Academic sources performed surprisingly poorly on transparency (JAMA Benchmark 2.36). Overall, content complexity typically exceeded the recommended 6th- to 8th-grade reading levels for public health information.</div></div><div><h3>Conclusions</h3><div>Patient information needs are tumor-specific, reflecting distinct clinical anxieties regarding survival versus function. A significant discordance exists in the current online landscape: highly transparent government resources are often too complex, while accessible commercial content lacks transparency. Clinicians must guide patients toward resources that balance accuracy with readability to ensure informed decision-making.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100224"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145939563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Aqueous humor derived cell free DNA: A minimally invasive tool for RB tumor mutation detection","authors":"Rachna Meel ,&nbsp;Renu Singh ,&nbsp;Sahil Agarwal ,&nbsp;Pooja Chauhan ,&nbsp;Seema Kashyap ,&nbsp;Lata Singh ,&nbsp;Arundhati Sharma","doi":"10.1016/j.ajoint.2025.100211","DOIUrl":"10.1016/j.ajoint.2025.100211","url":null,"abstract":"<div><h3>Purpose</h3><div>Retinoblastoma (RB) is the most common intraocular tumor in the pediatric age group. Aqueous humor (AH)-derived cell free (cf)DNA has recently emerged as a reliable surrogate for tumor DNA. The present study aimed to strengthen the results of our previous study on the feasibility of genetic analysis in AH derived cfDNA and compare it to peripheral blood and tumor tissue of RB patients.</div></div><div><h3>Methods</h3><div>Paired samples of AH, peripheral blood, and tumor tissue were collected in five patients with advanced intraocular retinoblastoma (ICRB Group D/E) undergoing primary enucleation. DNA from blood leukocytes, tumor tissue and AH cf DNA was isolated using standard protocols and QIAamp® Circulating Nucleic Acid kits respectively and subjected to next generation sequencing to identify <em>RB1</em> mutations across different sample sources.</div></div><div><h3>Results</h3><div>The mean concentration of AH cf DNA from five group E retinoblastoma eyes was 75.56 ng/ul (range 41.3-&gt;120ng/ul). Loss-of-function <em>RB1</em> mutations identified were RB1c.1363C&gt;<em>T</em> (p.Arg455Ter), c.1735C&gt;<em>T</em>(p.Arg579Ter), <em>RB1</em>cc.381–2A&gt;<em>G</em> (3′ splice site), RB1c.2330del (p.Pro777LeufsTer33) in four males and RB1c.1072C&gt;<em>T</em> (p.Arg358Ter) in one female patient. Concordance in the mutations identified in AH and tumor tissue was seen while no mutations were identified from blood (leucocytes) in any of the cases. Histopathology of the tumor tissue revealed poor differentiation in four and moderate differentiation in one eye with varying degrees of necrosis and calcification.</div></div><div><h3>Conclusion</h3><div>Our study supports previous evidence of the role of AH cfDNA in detecting tumor-specific mutations in RB with high concordance with tumor tissue, thus furnishing incremental evidence on the feasibility of genetic analysis of AH derived cfDNA of RB patients in an Indian tertiary-care context.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100211"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145798447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing glaucoma progression detection: Simulated evaluation of 24-2 and 10-2 visual field testing strategies using real-world data","authors":"Bethany E Higgins ,&nbsp;Giovanni Montesano ,&nbsp;Masaki Tanito ,&nbsp;Shiro Mizoue ,&nbsp;Kazuhiko Mori ,&nbsp;Katsuyoshi Suzuki ,&nbsp;Takehiro Yamashita ,&nbsp;Kazunori Hirasawa ,&nbsp;Nobuyuki Shoji ,&nbsp;David Crabb ,&nbsp;Ryo Asakoa","doi":"10.1016/j.ajoint.2026.100231","DOIUrl":"10.1016/j.ajoint.2026.100231","url":null,"abstract":"<div><h3>Objective/Purpose</h3><div>To evaluate whether alternating 24–2 and 10–2 visual field (VF) testing improves overall glaucoma progression detection compared with 24–2 or 10–2 alone, and to determine whether this enhances detection within the macular region, using simulated series derived from real-world data.</div></div><div><h3>Design</h3><div>Retrospective, observational simulation study.</div></div><div><h3>Subjects, Participants, and/or Controls</h3><div>Retrospective VF data from 426 eyes were used for simulations.</div></div><div><h3>Methods, Intervention, or Testing</h3><div>Pointwise slopes were estimated from 24–2 and 10–2 VFs using hierarchical modelling with location-level random intercepts/slopes. Eye-specific residuals formed noise templates. These were combined with estimated slopes to simulate 426 dense VF series (every 3-months), representing stable (slope=0) and progressing eyes. Four strategies were created and tested over 24-months: (A) simultaneous 24–2 and 10–2 at every visit, (B) 24–2 alone, (C) 10–2 alone and (D) alternating 24–2 and 10–2. Progression was determined using hierarchical modeling and strategy detection (hit) rate was assessed via cluster-adjusted ROC analyses, reporting sensitivity at 95% specificity and partial AUCs.</div></div><div><h3>Results</h3><div>At 12-months, sensitivity at 95% specificity was 79%, 67%, 67% and 63% (A–D); corresponding partial AUCs were 0.033 (A) and 0.025 (B–D). A had a significantly higher AUC than B, C, and D at 12-months (<em>p</em> &lt; 0.001 each). B exceeded C at 12 months (<em>p</em> = 0.015), while D and B did not differ (<em>p</em> = 0.570) and D exceeded C (<em>p</em> = 0.016). By 24-months, sensitivities converged for A and B (87% each), with 79% (C) and 80% (D). Similar results were seen in 244 eyes defined as early-stage glaucoma at baseline. For 284 eyes defined as macular progressors at 12-months, B detected 80% and D 81% of eyes identified by 10–2.</div></div><div><h3>Conclusions</h3><div>Alternating 24–2 and 10–2 testing provided no clear benefit over 24–2 alone, which remained robust for progression detection. Simultaneous testing is optimal but not clinically practical.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100231"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing glaucoma care in Ghana: a mixed-methods study of optometrists' practices, challenges, and capacity needs","authors":"Irene Nkansa-Kyeremateng ,&nbsp;Albert Kwadjo Amoah Andoh ,&nbsp;Eldrick Adu Acquah ,&nbsp;Isaiah Osei Duah Junior ,&nbsp;Beatrice Okrah ,&nbsp;Kwadwo Owusu Akuffo","doi":"10.1016/j.ajoint.2025.100204","DOIUrl":"10.1016/j.ajoint.2025.100204","url":null,"abstract":"<div><h3>Purpose</h3><div>To assess the practice patterns, challenges, and improvement strategies in glaucoma care among optometrists in Ghana.</div></div><div><h3>Design</h3><div>A cross-sectional study using a mixed-methods approach.</div></div><div><h3>Methods</h3><div>Quantitative data on demographics, clinical practices, associated factors, perceived barriers, and recommendations for improved glaucoma care were collected from 293 licensed optometrists registered with the Ghana Optometric Association using a structured questionnaire (distributed both in-person and online via Microsoft Forms), yielding a response rate of 44.1%. Qualitative insights were obtained through semi-structured interviews with nine optometrists, selected using purposive stratified sampling.</div></div><div><h3>Results</h3><div>Most respondents had formal training in glaucoma care through didactic courses and practical training in their completed degree programs. However, over one-third lacked standardized clinical care protocols. Clinical care centered on intraocular pressure measurement and optic nerve evaluation, with medical therapy as the primary treatment. Referrals were commonly triggered by poor treatment response (mainly based on IOP measurement) or advanced disease. The existence of clearly defined clinical protocols demonstrated a statistically significant association with increased provision of glaucoma care among optometrists (<em>p</em> = 0.008). Thematic analysis identified key challenges to the provision of glaucoma care, including patient noncompliance, financial barriers, and limited access to diagnostic tools. Recommended strategies for improvement included enhanced training, improved diagnostic infrastructure, and increased public education.</div></div><div><h3>Conclusion</h3><div>Optometrists play a vital role in glaucoma care in Ghana. However, gaps in guideline awareness, protocol use, and diagnostic capacity remain. Strengthening training, equipping clinics, and promoting public education are essential to improving glaucoma care nationwide.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100204"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145694956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Association between sociodemographic factors and visual impairment at initial presentation: A SOURCE data repository analysis","authors":"Xi Dai ,&nbsp;Muhammad Jehanzeb Khan ,&nbsp;Ahmed Sabit ,&nbsp;Mustafa Iftikhar ,&nbsp;Jiangxia Wang ,&nbsp;Jessica Chow ,&nbsp;Joshua D. Stein ,&nbsp;Fasika A. Woreta","doi":"10.1016/j.ajoint.2025.100210","DOIUrl":"10.1016/j.ajoint.2025.100210","url":null,"abstract":"<div><h3>Purpose</h3><div>To assess sociodemographic factors associated with the extent of VI at initial presentation from four of the most common causes of vision loss in the US.</div></div><div><h3>Design</h3><div>Retrospective cohort analysis</div></div><div><h3>Participants</h3><div>Patients ≥ age 40 in the Sight Outcomes Research Collaborative (SOURCE) Ophthalmology Data Repository presenting with cataract, glaucoma, diabetic retinopathy, or age-related macular degeneration ≥ 1 eye between Jan 1, 2012 and Dec 31, 2020.</div></div><div><h3>Methods</h3><div>VI was categorized into none/mild (20/40 or better), moderate (&lt; 20/40 to &gt; 20/200), and severe (20/200 or worse) based on best recorded better acuity in the better seeing eye at initial presentation. A multinomial logistic regression model was used to evaluate risk factors associated with VI at initial presentation.</div></div><div><h3>Main Outcome Measures</h3><div>Odds of moderate and severe VI at initial visit.</div></div><div><h3>Results</h3><div>We included 331,473 patients across seven participating sites. The mean (SD) age was 67.5 (10.5) years and 58.8% were female. Across the four disease categories, compared with White patients, Black, Asian, Hispanic, and Native American patients had higher odds of moderate VI (aOR 1.32, 1.35, 1.62, 1.51) and severe VI (aOR 1.59, 1.07, 1.50, 1.56), respectively. Non English speaking patients had higher odds of presenting with moderate and severe VI (aOR 1.17 and 2.25). Patients living in less affluent communities were also more likely to present with moderate (aOR 1.12, 1.22, 1.29, 1.12) and severe VI (aOR 1.30, 1.37, 1.47, 1.42) across the four disease categories. For Native American patients with severe VI, the adjusted odds ratio was 1.56 (95% CI 1.48 to 1.65).</div></div><div><h3>Conclusions</h3><div>These findings highlight the importance of targeted public health strategies, including language-concordant care to address disparities in vision health and promote earlier diagnosis and treatment for vulnerable populations in the US.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100210"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145798362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What patients search: Analyzing Google queries on laser refractive surgery and information transparency","authors":"Siddharth Gandhi ,&nbsp;Anne Cao ,&nbsp;Michael Balas ,&nbsp;Davin Johnson","doi":"10.1016/j.ajoint.2025.100223","DOIUrl":"10.1016/j.ajoint.2025.100223","url":null,"abstract":"<div><h3>Objective/Purpose</h3><div>To analyze the most common patient questions about laser refractive surgery on Google and to evaluate the transparency of the corresponding online information.</div></div><div><h3>Design</h3><div>Cross-sectional study.</div></div><div><h3>Subjects, Participants, and/or Controls</h3><div>The first 100 Google “People Also Ask” (PAA) questions were extracted for each of five search terms (“Laser-assisted in situ keratomileusis”, “LASIK”, “Photorefractive Keratectomy”, “PRK”, and “Laser eye surgery”). After deduplication, the study sample comprised 189 unique PAA questions and 284 unique source websites.</div></div><div><h3>Methods, Intervention, or Testing</h3><div>Data were collected on September 27, 2024, using an automated script. Questions were categorized by topic and type using Rothwell’s classification system. Source websites were assessed for transparency using the Journal of the American Medical Association (JAMA) Benchmark Criteria. A repeat data extraction was performed on November 17, 2025, to assess temporal stability.</div></div><div><h3>Results</h3><div>Of the 189 unique questions analyzed (88 LASIK, 60 PRK, 41 laser eye surgery), most focused on evaluating the procedures (26 %), recovery timelines (17 %), and indications/management (14 %). Fact-based questions were the most common (43 %). Among the 284 unique websites, medical practice websites were the most frequent (75 %) but had amongst the lowest mean JAMA scores (1.64). Commercial websites had the highest JAMA scores (3.42). LASIK-related searches had the highest mean JAMA score (2.01), followed by laser eye surgery (2.00) and PRK (1.67). The repeat data extraction on November 17, 2025, demonstrated low temporal stability, with only 9 % to 18 % of questions and 18 % to 27 % of websites recurring.</div></div><div><h3>Conclusions</h3><div>At the primary sampled time point, a transparency deficit characterized the online information most frequently encountered by patients. Although medical practice websites were the most visible sources, they often lacked transparency compared to other site types. These findings suggest that clinicians and academic institutions should prioritize the transparency of their online educational content to better support informed patient decision making.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100223"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145939564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of artificial intelligence for ophthalmic emergencies in emergency departments: A narrative review","authors":"Victor T.T. Chan ,&nbsp;Noel C.Y. Chan","doi":"10.1016/j.ajoint.2026.100225","DOIUrl":"10.1016/j.ajoint.2026.100225","url":null,"abstract":"<div><h3>Purpose</h3><div>To review the current evidence on artificial intelligence (AI)-based diagnostic systems for ophthalmic emergencies and to evaluate their performance, clinical applicability, and implementation challenges in emergency department settings.</div></div><div><h3>Methods</h3><div>A narrative review was conducted with a systematic literature search of PubMed/MEDLINE and the Cochrane Library (January 2015–June 2025). Studies were included if they reported AI models for central retinal artery occlusion (CRAO), anterior ischemic optic neuropathy (AION), rhegmatogenous retinal detachment (RRD)/retinal breaks, acute angle closure (AAC), or infectious keratitis using readily available emergency-department imaging modalities (anterior-segment photographs, colour fundus photography, or retinal optical coherence tomography). AI architecture, imaging modality and specific devices, dataset size and source, level of validation, and reported performance metrics were extracted and critically appraised.</div></div><div><h3>Results</h3><div>Twenty-four studies met inclusion criteria. Deep learning models achieved high diagnostic performance in CRAO (area under curve [AUC] 0.96–0.99 from 2 studies), AION (AUC 0.97 from 1 study), RRD/ retinal break (AUC 0.888–1.00 from 7 studies), AAC (AUC 1.00 from 1 study), and infectious keratitis (AUC 0.65–0.997 from 13 studies). Promising results were also observed for smartphone-based fundus photography and portable OCT devices. However, most studies relied on retrospective, single-centre datasets with limited external or prospective validation.</div></div><div><h3>Conclusions</h3><div>AI demonstrates considerable potential to support rapid, accurate triage of sight-threatening ophthalmic emergencies, especially in resource-limited settings. Nevertheless, generalisability, data privacy, and integration into clinical pathways remain major barriers to routine adoption. Prospective multicentre trials with adequate external validation and privacy-by-design solutions are needed before widespread clinical implementation.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100225"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The association of diabetic retinopathy and visual impairment with falls and falls reporting injuries: Findings from the behavioral risk factor surveillance system","authors":"Neel Edupuganti ,&nbsp;Danny Alevy ,&nbsp;Tommy Bui ,&nbsp;George Liu ,&nbsp;Kevin Dobbin ,&nbsp;John E. Crews ,&nbsp;Atalie C. Thompson","doi":"10.1016/j.ajoint.2025.100221","DOIUrl":"10.1016/j.ajoint.2025.100221","url":null,"abstract":"<div><h3>Purpose</h3><div>To investigate the associations between visual impairment (VI) and diabetic retinopathy (DR) with the prevalence of falls, recurrent falls, and falls reporting injuries (FRIs).</div></div><div><h3>Design</h3><div>Cross-sectional analyses of associations between VI, DR, or both VI and DR with the prevalence of any fall, recurrent falls, and FRIs using data from 65,046 adults (45+ years) who completed the 2018 Behavioral Risk Factor Surveillance System (BRFSS) Diabetes Module.</div></div><div><h3>Subjects, participants, and/or controls</h3><div>Participants were categorized into four groups: no VI or DR (reference), DR only, VI only, and both VI and DR.</div></div><div><h3>Methods</h3><div>Separate multivariable logistic regression models, adjusted for sociodemographics and comorbidities, were used to analyze the association between VI, DR, or VI and DR with the odds of experiencing any fall, recurrent falls, or FRIs.</div></div><div><h3>Main Outcome Measures</h3><div>The association between VI, DR, or VI and DR with the odds of experiencing any fall, recurrent falls, or FRIs.</div></div><div><h3>Results</h3><div>Compared to adults with no VI or DR, those with VI reported 1.72 greater odds of experiencing any fall (adjusted odds ratio (aOR)=1.716, <em>p</em> &lt; 0.001), 1.82 greater odds of an increasing number of falls (aOR=1.824, <em>p</em> &lt; 0.001), and 2.08 greater odds of FRIs (aOR=2.076, <em>p</em> &lt; 0.001). Similarly, those with DR only also had substantially greater odds of experiencing any fall (aOR=1.741, <em>p</em> &lt; 0.001), an increasing number of falls (aOR = 1.735, <em>p</em> &lt; 0.001), and FRIs (aOR=1.658, <em>p</em> = 0.004). The interaction effect between VI and DR was not significant. Subjects with both VI and DR experienced 3.80 greater odds of any fall, 2.73 greater odds of increasing falls, and 2.79 greater odds of FRIs, which was greater than those with VI or DR alone or neither condition. Further, subjects aged 45–64 years had significantly higher odds of FRIs than those aged 65+.</div></div><div><h3>Conclusions</h3><div>Using data from 17 states and territories implementing the 2018 BRFSS Diabetes Module, we found that adults with both VI and DR experienced a greater likelihood of falls, recurrent falls, and FRIs than those with neither or either condition. However, the inability to infer temporal or causal relationships from this cross-sectional, self-reported analysis should be considered. These findings underscore the importance for practitioners to communicate fall risks among people with DR and VI and support policies promoting tailored interventions for people experiencing these specific conditions.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"3 1","pages":"Article 100221"},"PeriodicalIF":0.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identifying exudative recurrence in neovascular age-related macular degeneration using large language models","authors":"Mariapaola Giordano,&nbsp;Claudio Xompero,&nbsp;Carl-Joe Mehanna,&nbsp;Eric H. Souied","doi":"10.1016/j.ajoint.2025.100186","DOIUrl":"10.1016/j.ajoint.2025.100186","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the ability of a large language model (LLM) to identify exudative recurrence in neovascular age-related macular degeneration (nAMD) from optical coherence tomography (OCT) scans and compare its performance with retina specialists.</div></div><div><h3>Design</h3><div>Retrospective, single-center, observational study.</div></div><div><h3>Methods</h3><div>Two consecutive OCT scans of patients with nAMD under <em>pro re nata</em> (PRN) regimen were collected. Screen recordings of the complete OCT scan comparisons (previous vs. current) were analyzed by two retina specialists versus ChatGPT-4o after providing a standardized prompt. Main outcome measure was agreement between the LLM and the ophthalmologists in detecting exudative activity. Statistical measures included sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Cohen’s kappa for agreement.</div></div><div><h3>Results</h3><div>A total of 102 eyes were included. Among 71 confirmed recurrences, ChatGPT-4o correctly identified 63 and missed 8; of 31 non-recurrences, it correctly identified 10 and misclassified 21 as positive. Overall accuracy was 71.57 %, with sensitivity of 88.73 % and specificity of 32.26 %. Agreement with ophthalmologists was fair (Cohen’s <em>k</em> = 0.238).</div></div><div><h3>Conclusion</h3><div>ChatGPT-4o demonstrated strong sensitivity in detecting exudative recurrences in nAMD, but limited specificity. Given the small, single-center cohort, the lack of external or same-dataset validation, these results should be considered preliminary, but highlights as LLM-assisted analysis, while not yet capable of replacing clinical expertise, may serve as a valuable adjunct for screening exudative changes in nAMD. Further refinements in the LLM could improve specificity and clinical utility.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"2 4","pages":"Article 100186"},"PeriodicalIF":0.0,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145416833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ability of the optical coherence tomography parameters to identify retinal nerve fibre layer defect in Indian eyes with mild glaucoma","authors":"Flonny Sunny,&nbsp;Premanand Chandran,&nbsp;Liji Joy,&nbsp;Ganesh V Raman","doi":"10.1016/j.ajoint.2025.100196","DOIUrl":"10.1016/j.ajoint.2025.100196","url":null,"abstract":"<div><h3>Purpose</h3><div>To compare the diagnostic ability of thickness, deviation and clock-hour map of spectral domain optical coherence tomography (OCT) to detect retinal nerve fibre layer (RNFL) defect in mild glaucoma.</div></div><div><h3>Design</h3><div>Cross sectional study.</div></div><div><h3>Methods</h3><div>Eyes with mild glaucoma and well defined photographic RNFL defect were included as cases and eyes with normal optic nerve head were included as controls. Fundus photograph and spectral domain OCT optic disc cube scan was performed in all the eyes. Thickness map, deviation map and clock-hour map of OCT were assessed for the presence of RNFL defect. Image J was used to calculate the width of the RNFL defect.</div></div><div><h3>Results</h3><div>127 eyes were recruited in each group. OCT RNFL thickness map was able to identify the RNFL defect in all the 127 (100 %) eyes, whereas deviation map was able to identify in 122 (96 %) eyes and clock-hour map in 114 (90 %) eyes. Average width of RNFL defect was 21⁰ (± 5.9). The width of the RNFL defect was &lt;20° in all the eyes in which RNFL defect was missed by both deviation and clock-hour map. Narrow angular width of RNFL defect was the only risk factor for missing RNFL defect in both deviation and clock-hour map. Among the control eyes thickness map showed RNFL thinning in 3 (2.4 %) eyes, deviation map in 6 (4.7 %) eyes and clock-hour map in 7 (5.5 %) eyes.</div></div><div><h3>Conclusions</h3><div>Thickness map performs better than deviation map and clock-hour map in identifying RNFL defect in eyes with mild glaucoma.</div></div>","PeriodicalId":100071,"journal":{"name":"AJO International","volume":"2 4","pages":"Article 100196"},"PeriodicalIF":0.0,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145466038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}