36 papers
This mixed-method study reveals that no valid, standardized intelligence tests or norm-groups currently exist for adults with visual impairment, leading healthcare professionals to rely on verbal subtests and ad-hoc adaptations while highlighting an urgent need for tailored assessment tools.
This study demonstrates that a multimodal machine learning approach, specifically a multi-layer perceptron integrating clinical, structural, and functional data, achieves superior diagnostic performance for glaucoma detection in a West African cohort compared to traditional models and individual metrics, highlighting its potential for resource-constrained settings.
This prospective comparative study found that AI chatbots, particularly ChatGPT and DeepSeek, provide reliable and high-quality information on myopia suitable for helping parents and patients understand the condition, while Gemini's responses were rated slightly lower but still adequate.
This paper presents a reproducible, step-by-step tutorial on using transfer learning with the ResNet152V2 model to achieve 98.8% accuracy in classifying thirteen distinct eye diseases based on visual symptoms, providing a practical educational resource for healthcare professionals and learners.
This study demonstrates that axial length significantly biases retinal vascular metrics derived from fundus images, but these systematic errors can be effectively corrected using the Bennett-Littmann formula to ensure accurate biomarker development for systemic diseases.
This study demonstrates that short-term narrowband blue light exposure significantly reduces axial length more effectively in the morning than in the evening, thereby altering the eye's natural diurnal rhythm, whereas adding myopic defocus to the exposure provides no additional short-term benefit.
This one-year cross-sectional study of 296 type 2 diabetes patients at a tertiary hospital in Northern Tanzania reveals that dyslipidemia, particularly elevated serum triglycerides, is significantly and independently associated with the presence of diabetic macular edema.
The paper introduces CausalFund, a causality-inspired framework that enhances the domain generalization of AI models for glaucoma and diabetic retinopathy screening by disentangling disease-specific features from spurious image factors, thereby enabling reliable diagnosis across diverse clinical and low-resource settings using portable devices.
Using whole-genome sequencing data from the All of Us Research Program, this study identifies a rare 8,732 base pair deletion upstream of the PITX2 gene that disrupts multiple enhancers and confers a seven-fold increased risk for primary open-angle glaucoma, demonstrating the value of structural variant analysis in uncovering high-impact genetic factors missed by conventional genotyping.
This study reveals that L-ergothioneine accumulates in ocular tissues to protect against oxidative damage and identifies significantly lower levels of this antioxidant in patients with neovascular age-related macular degeneration, suggesting a strong link between low ergothioneine and the disease.
This real-world study of over 80,000 encounters across 33 practices demonstrates that an integrated remote physiologic monitoring and principal care management program using home-based retinal function testing effectively detects asymptomatic disease progression in chronic retinal conditions, leading to timely clinical interventions with high patient adherence.
This study demonstrates that a Vision Transformer-based deep learning model can accurately predict future visual field indices from a single baseline OCT-RNFL scan, showing robust performance across different OCT devices and offering a novel tool for anticipating glaucoma progression.
This retrospective real-world analysis of 1,000 eyes treated with the aflibercept biosimilar aflibercept-ayyh (PAVBLU) in US retina practices demonstrated stable visual acuity in previously treated patients, improved vision in treatment-naive patients, and a safety profile consistent with the reference drug, supporting its use as an effective alternative for retinal disorders.
This multi-cohort exome-wide meta-analysis of African ancestry individuals identifies rare coding variants in genes such as SRF, BLTP3A, METTL2A, and KRT10 as potential ancestry-specific contributors to primary open-angle glaucoma, highlighting the critical need to study underrepresented populations to better understand the disease's etiology.
This cross-sectional study utilized an interpretable random forest machine learning model to demonstrate that ocular biometric factors, particularly age, axial length, and the anterior chamber depth-to-axial length ratio, exhibit strong nonlinear associations with cataract development in highly myopic eyes, outperforming systemic laboratory factors in predictive significance.
This study introduces a multimodal AI framework called COMET that integrates large-scale electronic health records with proteomic data to rationally prioritize and validate novel protein biomarkers for diabetic retinopathy, demonstrating superior predictive performance and biological relevance compared to single-modality approaches.