EyePub Date : 2026-05-08DOI: 10.1038/s41433-026-04491-4
Kosar Esmaili, Rohith Erukulla, Ikesinachi Osuorah, Mehdi Aminizade, Kasra Cheraqpour, Amir Rahdar, Emine Esra Karaca, Ayşe Nilay Özgür, Özlem Evren Kemer, Seyed Ali Tabatabaei, Reza Mirshahi, Albert Y Cheung, Natalia Quiroz-Casian, Zahra Bibak-Bejandi, Seyed Farzad Mohammadi, Raghuram Koganti, Siamak Yousefi, Mohammad Soleimani
{"title":"Artificial intelligence-assisted diagnosis and subtype differentiation of infectious keratitis.","authors":"Kosar Esmaili, Rohith Erukulla, Ikesinachi Osuorah, Mehdi Aminizade, Kasra Cheraqpour, Amir Rahdar, Emine Esra Karaca, Ayşe Nilay Özgür, Özlem Evren Kemer, Seyed Ali Tabatabaei, Reza Mirshahi, Albert Y Cheung, Natalia Quiroz-Casian, Zahra Bibak-Bejandi, Seyed Farzad Mohammadi, Raghuram Koganti, Siamak Yousefi, Mohammad Soleimani","doi":"10.1038/s41433-026-04491-4","DOIUrl":"https://doi.org/10.1038/s41433-026-04491-4","url":null,"abstract":"<p><strong>Background: </strong>Infectious keratitis (IK) is a major cause of corneal blindness world-wide, and prompt identification of IK and its etiologic subtype is essential for appropriate management. We developed deep learning (DL) models to detect IK and differentiate common subtypes from slit-lamp photographs.</p><p><strong>Methods: </strong>In this retrospective study, slit-lamp photographs were collected from patients presenting to the emergency department of Farabi Eye Hospital (2014-2021) with bacterial keratitis (BK), fungal keratitis (FK), Acanthamoeba keratitis (AK), or herpes simplex keratitis (HSK), along with healthy controls and corneal scars. A total of 13,953 images were included. Three DL classifiers were trained: Model 1 (IK vs. normal), Model 2 (healthy vs. corneal scar vs. IK [pooled subtypes]), and Model 3 (BK vs. FK vs. AK vs. HSK).</p><p><strong>Results: </strong>Model 1 achieved 99.9% accuracy for IK vs. normal (ROC-AUC 0.999). In five-fold cross-validation, Model 2 achieved mean accuracy 0.975 (95% CI 0.955-0.996), macro-F1 0.970 (95% CI 0.945-0.995), and macro-average AU-ROC 0.998 (95% CI 0.995-1.000). For subtype classification (Model 3), overall accuracy was 81.6% with balanced recall 83.3%; class accuracies were 88% (BK), 71% (FK), 72% (AK), and 93% (HSK) with ROC-AUCs 0.90-0.98. External validation of Model 3 (Ankara City Hospital; 665 images from 96 patients) showed accuracy 92.5%, macro-F1 93%, macro-average AUROC 0.996, and sensitivities of 95.2% (AK), 92.0% (BK), 85.5% (FK), and 99.5% (HSK).</p><p><strong>Conclusions: </strong>DL models applied to slit-lamp photographs showed high performance for IK detection and clinically relevant differentiation of IK from corneal scars and among major IK subtypes, with external validation supporting generalisability.</p>","PeriodicalId":12125,"journal":{"name":"Eye","volume":" ","pages":""},"PeriodicalIF":3.2,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147856143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
EyePub Date : 2026-05-06DOI: 10.1038/s41433-026-04500-6
Chloe Wagstaff, Michael Williams, Jonathan Jackson, Rebecca McCracken, Sara McCracken, Gerard J Gormley
{"title":"Through their eyes': A qualitative study on the impact of virtual reality on parents' understanding of visual impairment.","authors":"Chloe Wagstaff, Michael Williams, Jonathan Jackson, Rebecca McCracken, Sara McCracken, Gerard J Gormley","doi":"10.1038/s41433-026-04500-6","DOIUrl":"https://doi.org/10.1038/s41433-026-04500-6","url":null,"abstract":"<p><strong>Background: </strong>Children born with visual impairment perceive the world differently, which can lead to developmental delays that are often difficult for parents to understand. This uncertainty can leave them unsure of how best to support their child. Virtual Reality (VR) offers simulated experiences and is increasingly being explored as a tool to enhance empathy and understanding - particularly through Point of View (PoV) simulations. PoV VR experiences have the potential to provide personalised insights into a child's visual impairment (VI). This study addresses the research question: What is the lived experience of parents and carers engaging in a VR simulation replicating their child's visual condition?</p><p><strong>Methods: </strong>We conducted a qualitative study to explore the lived experiences of parents and carers using a PoV VR learning experience. Semi-structured interviews, guided by a phenomenological approach, were used to gather in-depth insights. Ten parents/carers of children with VI were recruited and interviewed. Transcripts were analysed using template analysis.</p><p><strong>Results: </strong>Four themes emerged that captured the depth of participants' experiences: (1) 'It all makes sense now': VR allowing carers to 'peer' through their child's eyes; (2) 'A flood of emotions'; (3) 'Seeing into their child's future' and (4) 'Empowerment and advocacy' CONCLUSIONS: PoV VR enhances empathy and understanding of children's visual abilities, empowering parents and carers to advocate more effectively for their needs. This immersive tool offers deeper insights into the lived experience of VI, underscoring its value in improving support for children with VI and their families.</p>","PeriodicalId":12125,"journal":{"name":"Eye","volume":" ","pages":""},"PeriodicalIF":3.2,"publicationDate":"2026-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147835674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
EyePub Date : 2026-05-06DOI: 10.1038/s41433-026-04499-w
Siyuan Jabelle Lu, Shenouda Girgis, Peter Shah, Graham A Lee
{"title":"Patient experience of and barriers to the eye examination.","authors":"Siyuan Jabelle Lu, Shenouda Girgis, Peter Shah, Graham A Lee","doi":"10.1038/s41433-026-04499-w","DOIUrl":"https://doi.org/10.1038/s41433-026-04499-w","url":null,"abstract":"<p><strong>Background: </strong>Ophthalmic examination is central to the diagnosis and monitoring of eye disease. However, there is a paucity of qualitative studies about the patient's experience. The Patient Experience of Eye Examination eValuation Study (PEEEVS) was designed to evaluate patient experiences of key examination-related components of routine ophthalmic care.</p><p><strong>Methods: </strong>PEEEVS employs a cross-sectional, mixed methods design combining quantitative visual analogue scales (VAS) with qualitative semi-structured interviews. Data from 203 patients (M:F - 101:102) were analysed with respect to their experiences of topical anaesthesia, optical coherence tomography (OCT) imaging, tonometry, slit lamp examination, and mydriasis.</p><p><strong>Results: </strong>Quantitative analysis indicated predominantly favourable responses, with median VAS scores ranging between 88 and 89 for all tests. In contrast, qualitative findings identified specific areas of concern, including challenges with maintaining proper positioning during slit lamp examination and OCT, particularly for individuals with pre-existing neck/back pain, larger body habitus, or advanced pregnancy.</p><p><strong>Conclusions: </strong>While participant experience of most monitoring tests was good, specific examination aspects can impact patient comfort and potentially affect long-term adherence to ocular monitoring. The study highlights the need for targeted considerations, such as ergonomic improvements and enhanced patient communication strategies to alleviate discomfort and support sustained engagement in ophthalmic care.</p>","PeriodicalId":12125,"journal":{"name":"Eye","volume":" ","pages":""},"PeriodicalIF":3.2,"publicationDate":"2026-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147835668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
EyePub Date : 2026-05-06DOI: 10.1038/s41433-026-04482-5
Laura Elizabeth Boddy, ShinYi Woo, Pollyanna Murch, Dalia Said, Harminder Singh Dua
{"title":"Evaluation of an allied health professionals-led keratoconus service in a tertiary UK centre.","authors":"Laura Elizabeth Boddy, ShinYi Woo, Pollyanna Murch, Dalia Said, Harminder Singh Dua","doi":"10.1038/s41433-026-04482-5","DOIUrl":"https://doi.org/10.1038/s41433-026-04482-5","url":null,"abstract":"<p><strong>Background: </strong>Keratoconus is an ectatic corneal disorder, which requires regular monitoring for early detection of progression and prompt treatment with collagen cross-linking (CXL) to stabilise the condition. This conflicts with the demand on corneal specialists to review acute referrals into the corneal service, delaying diagnosis and treatment. In this study, we examined the outcomes of an Allied Health Professional (AHP)-led keratoconus service for routine workup, monitoring, performing CXL and follow-up thereafter.</p><p><strong>Methods: </strong>Refraction, topography and CXL operative details were taken from electronic patient records. Statistical analysis was performed in R<sup>®</sup> (R Foundation for Statistical Computing, Vienna, Austria). 82 eyes had CXL performed by a nurse practitioner (NP) and 42 eyes by ophthalmologists. Pre- and post-operative evaluation were done by optometrists.</p><p><strong>Results: </strong>In total, 271 eyes of 177 patients were included. 92 patients (128 eyes) were consented and underwent CXL. 85 patients (143 eyes) where CXL was not indicated, were monitored routinely. All patients had baseline refraction and topography. Patients were reviewed at 12 and 24 months. All had topography at every visit. 71/72 eyes (98.61%) had refraction at 12 months and 57/58 eyes (91.28%) had refraction at 24 months. Patients who were operated on by NP found improvements in visual acuity at 12 months and 24 months (p < 0.05, Mann-Whitney U Test).</p><p><strong>Conclusion: </strong>The AHP-led keratoconus service delivered satisfactory outcomes and can be an efficient means of dealing with these patients, complementing specialist corneal services in the NHS.</p>","PeriodicalId":12125,"journal":{"name":"Eye","volume":" ","pages":""},"PeriodicalIF":3.2,"publicationDate":"2026-05-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147835685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}