Search Results (334)

Background and Objectives: Pseudoexfoliative glaucoma (PEXG) is a secondary glaucoma caused by the deposition of abnormal flaky protein-like material inside the eye, leading to blockage of the drainage system, increased intraocular pressure (IOP), and optic nerve damage. This study aimed to evaluate the effects of latanoprost and timolol–dorzolamide therapy on the concentrations of interleukin (IL)-1β, IL-6, IL-8, IL-10, IL-12p70, and tumor necrosis factor-alpha (TNF-α) in the aqueous humor and serum samples of patients with PEXG. Materials and Methods: This prospective observational study included 60 patients allocated to three age-matched groups (n = 20 per group). Group 1 received latanoprost therapy, Group 2 received timolol–dorzolamide therapy, and Group 3 comprised patients with cataracts without PEXG who were not receiving antiglaucoma medication (control group). All participants underwent a comprehensive ophthalmological examination, including visual acuity assessment, specular microscopy, IOP measurement, and intraocular lens (IOL) measurement. Peripheral venous blood was taken from each patient before the surgical procedure and aqueous humor samples were collected during cataract surgery. Concentrations of IL-1β, IL-6, IL-8, IL-10, IL-12p70, and TNF-α were determined using a Cytometric Bead Array and flow cytometry. Results: No significant differences were observed in the aqueous humor and serum concentrations of IL-6 and IL-8 between the treatment groups, while IL-1β, IL-10, IL-12p70, and TNF-α did not reach detection. A significant positive correlation was found between IOL and aqueous humor IL-6 concentration in group 1 and between IL-6 and IL-8 serum concentration and IOP in controls. Conclusions: Latanoprost and timolol–dorzolamide therapy were not associated with significant differences in aqueous humor or serum IL-6 and IL-8 concentrations in patients with PEXG. IL-1β, IL-10, IL-12p70, and TNF-α were not detectable in the analyzed samples. These findings support the presence of low-grade inflammation in patients with PEXG that does not differ according to the therapy administered. Corneal endothelial alterations are disease-related rather than treatment-induced. Full article

Eye diseases, particularly glaucoma and cataracts, are the leading causes of visual impairment, compromising quality of life. Automatic classification of these diseases can lead to more accurate and timely diagnosis, thereby limiting their progression and complications. In recent years, advances in Deep Learning (DL) have shown promising results in the study of images in ophthalmology. However, traditional DL models are based on a centralised approach to data, sharing sensitive patient information and compromising privacy; moreover, the models are often difficult to interpret. In this paper we propose a method aimed to solve the issues of privacy and transparency in decision-making related to eye diseases detection and localisation. As a matter of fact, we consider Federated Learning (FL), an approach based on data decentralisation that enables collaborative learning between different clients and sends only the model weights to the central server. In this way, sensitive patient data are not shared, ensuring security and privacy. With regard to eye disease classification we exploit a Vision Transformer, which allows global relationships within retinal images to be highlighted, improving representation capabilities compared to traditional convolutional architectures. Furthermore, the proposed method also aims to make the model explainable using explainability techniques, in this way we make diagnostic decisions transparent. The experimental analysis shows an accuracy of 0.8480, a precision of 0.8645, a recall of 0.8477, showing the effectiveness of the proposed method on eye disease detection. Full article

Background: Obesity has become a major global public health challenge, affecting individuals across all age groups and contributing to a wide range of systemic disorders. Beyond its well-established associations with cardiovascular and metabolic diseases, obesity is increasingly recognized as a chronic inflammatory condition that influences the structure and function of multiple organs, including the eye. Methods: A comprehensive literature review was conducted between November 2025 and June 2026 using PubMed, Web of Science, and Google Scholar. Priority was given to recent studies evaluating the impact of obesity and obesity-related metabolic disturbances on ocular health. Results: Available evidence indicates that obesity is associated with an increased risk of several ophthalmic conditions, including age-related macular degeneration, diabetic retinopathy, glaucoma, cataracts, and retinal vein occlusion. Multiple anthropometric measures of adiposity have been positively associated with diabetic retinopathy risk. Although findings regarding glaucoma remain heterogeneous, metabolic syndrome and its components, particularly impaired glucose metabolism, appear to increase glaucoma susceptibility. Obesity has also been linked to dry eye disease through inflammatory and tear film alterations. Conclusions: Obesity exerts significant effects on ocular health through complex metabolic, inflammatory, and vascular pathways. Recognition of obesity as a modifiable risk factor for eye disease may facilitate earlier detection, targeted prevention strategies, and improved multidisciplinary management. Further longitudinal and mechanistic studies are needed to clarify causal relationships and determine whether effective obesity treatment can reduce the burden of vision-threatening ocular disorders. Full article

Background/Objectives: Primary open-angle glaucoma (POAG) is a complex multifactorial optic neuropathy involving genetic, vascular, oxidative, inflammatory, and neurodegenerative mechanisms. Despite advances in genome-wide studies, the contribution of genetic variants remains incompletely characterized in underrepresented Latin American populations. This study aimed to characterize the genetic landscape of POAG in a Colombian cohort by identifying previously reported glaucoma-associated variants, rare candidate variants, and pharmacogenomic markers and integrating these findings into biologically relevant pathways. Methods: An exploratory descriptive study was conducted in 21 Colombian patients with confirmed POAG. Whole-exome sequencing (WES) was performed at an average sequencing depth of approximately 100×. Variants were quality-filtered, functionally annotated, and prioritized within 446 POAG-associated genes retrieved from DisGeNET. Previously reported glaucoma-associated variants and rare candidate variants were identified, while pharmacogenomic variants related to latanoprost and timolol response were evaluated using ClinPGx/PharmGKB. Identified genes were classified according to major biological pathways relevant to glaucoma pathophysiology. Results: Of the 446 POAG-associated genes, 381 were detected in the patients’ exomes. A total of 10,220 molecular variants were identified, of which 1,187 synonymous variants were excluded, leaving 9,033 variants for downstream analysis. Among these, 955 were non-synonymous SNVs, including 26 variants previously reported in association with glaucoma and 929 potentially novel coding variants. Previously reported variants included loci in SIX6, LOXL1, CYP1B1, NOS3, and SOD2. Two rare candidate variants (minor allele frequency <1%) were identified in FMNL2 and C3. Pharmacogenomic variants in PTGS1, ADRB1, and ABCC4 with potential implications for response to latanoprost or timolol were also detected. Functional integration highlighted pathways involving oxidative stress, extracellular matrix remodeling, vascular regulation, neurodegeneration, and inflammation. Conclusions: This exploratory analysis identifies known glaucoma-associated variants, rare candidate variants, and pharmacogenomic markers in Colombian patients with POAG. The findings support a multifactorial biological framework involving interconnected oxidative, structural, vascular, neurodegenerative, and inflammatory pathways. The FMNL2 and C3 variants represent candidates for further investigation, while the identified pharmacogenomic variants highlight the potential relevance of genomic profiling for personalized glaucoma management. Larger ancestry-informed case–control studies are required to validate these observations and determine their clinical significance. Full article

Background/Objectives: Glaucoma is a leading cause of irreversible blindness globally and must be detected early to preserve vision. Standard methods, such as automated perimetry and optical coherence tomography (OCT), are limited in their ability to detect early changes. Therefore, this systematic review evaluates the diagnostic and predictive performance of artificial intelligence (AI)-driven OCT-based models, including those based on deep learning, machine learning, and transformer architectures, and contextualizes their performance against the recognized limitations of standard automated perimetry and conventional OCT trend analysis. Methods: The PubMed, Ovid, and Google Scholar databases were comprehensively searched to identify relevant published research in English. Screening of observational, cohort, prospective, retrospective and diagnostic accuracy studies, as well as clinical trials, identified 13 studies. Their key outcomes, including sensitivity, specificity, and area under the curve (AUC), were extracted for synthesis. Results: AI-driven OCT-based models generally reported high diagnostic performance, with mean sensitivities and specificities exceeding 80%, as well as AUC values generally above 0.85. AI-driven OCT-based models were reported to detect glaucoma progression as early as 9 months before standard methods. Conclusions: The findings suggest that AI-driven OCT-based models may improve glaucoma detection by identifying subtle structural and functional changes earlier, which may help inform earlier intervention decisions. However, the majority of included studies relied on internal validation only, and prospective multicenter studies with external validation are required before AI-driven OCT-based models can be reliably implemented in routine glaucoma care. Full article

MicroPulse laser trabeculoplasty (MLT) lowers intraocular pressure (IOP) with subthreshold pulses that spare the surrounding tissue, yet its effect on the corneal endothelium is far less studied than that of selective laser trabeculoplasty, and almost unstudied in treatment-naïve eyes. We retrospectively reviewed 70 treatment-naïve glaucoma patients (one eye each) treated with a 577 nm yellow laser (Easyret^®) in micropulse mode (300 µm, 300 ms, 1000 mW, 360°). Endothelial cell density, mean cell area, hexagonality, the coefficient of variation in cell size, and central corneal thickness were recorded by specular microscopy (SP-1P, Topcon), and IOP by Goldmann applanation, at baseline and 1, 3, and 6 months. Linear mixed-effects models used every available observation, so no patient was excluded for a missed visit. IOP fell from 25.9 ± 2.5 mmHg to 20.6 and 21.0 mmHg at one and three months (−5.32 and −5.00 mmHg, roughly 20%; both p < 0.001), then returned to 24.8 mmHg by six months, and the proportion reaching a 20% reduction fell from 62.5% to 4.6%. Endothelial cell density, mean cell area, and central corneal thickness held steady, and every morphological change stayed within the device’s repeatability limits on equivalence testing. MLT lowered IOP early but transiently; over six months no clinically meaningful change in corneal endothelial morphology was detected. Full article

Diabetic retinopathy (DR) and glaucoma, globally leading causes of irreversible blindness, can be detected and managed early with timely screening. In low-resource settings, access to ophthalmological examination is limited by a variety of constraints, including scarcity of specialists, equipment costs, geographic distance from care centers, and limited patient awareness. This narrative review examines the principal barriers to early detection of glaucoma and DR and evaluates evidence-based strategies to overcome them, including low-cost portable ophthalmic testing tools, artificial intelligence (AI) analytics, risk-based resource allocation, teleophthalmology, and targeted patient education. In different populations, these strategies have been shown to meaningfully increase access to ophthalmologic screening and follow-up at a significantly lower cost. We describe these approaches, demonstrate previously successful initiatives, and propose various combined approaches adapted to the specific needs of the various low-resource settings. Full article

Glaucoma is a leading cause of irreversible blindness, and timely detection and monitoring are essential to prevent permanent visual loss. Artificial intelligence (AI) has shown strong potential for automated diagnosis, but progression prediction remains a more challenging and clinically significant task. This study proposes an explainable multimodal deep learning framework for both glaucoma detection and progression prediction using the Harvard Glaucoma Detection and Progression dataset. The framework integrates specific modality encoders for optical coherence tomography B-scans, retinal nerve fiber layer thickness (RNFLT) maps, and visual field (VF) features through a feature fusion module. For the glaucoma detection task, the multimodal model achieved near-perfect performance with an AUROC of 0.9989 and a sensitivity of 0.9944. For the more complex progression prediction task, the full-fusion model achieved an AUROC of 0.7836. Using an optimized validation threshold, the model provided a balanced clinical operating point with reasonable sensitivity and specificity. Ablation experiments revealed a complementary relationship between modalities; VF features provided the strongest standalone ranking signal, RNFLT maps contributed the highest specificity, and multimodal fusion yielded the best overall precision–recall performance, although not the highest AUROC. Gradient-weighted Class Activation Mapping analysis indicated that the model focused on clinically plausible localized retinal regions. These findings support explainable and multimodal AI for glaucoma progression-risk identification. Full article

Background/Objectives: Glaucoma is a leading cause of irreversible blindness, and it is hard to catch early be-cause it rarely causes symptoms until real damage has already occurred. Existing au-tomated detection methods still miss the subtle structural changes that show up before vision loss begins. This study presents an automated framework based on optic nerve structure and retinal biomarkers, aimed at enabling earlier and more consistent glau-coma screening. The framework combines segmentation and classification in a single deep learning pipeline, trained and tested on the ORIGA and REFUGE2 fundus image datasets. Each image is resized, denoised with median filtering, and contrast-enhanced through histogram equalisation and normalisation, then cropped down to the optic nerve region using central cropping and intensity-based localisation. Methods: A custom encod-er–decoder CNN segments the optic disc and cup, and from that segmentation, we ex-tract biomarkers ophthalmologists already rely on, including the vertical cup-to-disc ratio and disc/cup area measurements. A Lightweight Vision Transformer separately learns broader structural patterns across the retina, and the two feature sets are fused and passed through a SoftMax classifier. Results: Segmentation accuracy was strong: Dice scores of 0.9082 for the optic disc and 0.9994 for the optic cup, IoU scores of 0.8351 and 0.9988, and an overall mean Dice of 0.9538 and mean IoU of 0.9169. Classification performance held up well, too, with high F1-score, recall, accuracy, and precision across normal and glaucomatous cases. Conclusions: The combination of interpretable, clinically established biomarkers with transformer-based global feature learning gives the framework the ability to support automated glaucoma risk assessment and demon-strates promising performance for glaucoma detection using retinal fundus images. Full article

Retinal nerve fiber layer defect (RNFLD) is a critical early indicator of glaucoma. Existing RNFLD detection methods underutilize clinical anatomical priors, struggle with subtle or diffuse lesions, and rely heavily on high-quality annotations. To address these issues, this paper proposes Inexact Position-Wise Sector Contrastive Learning (IPSCL) for RNFLD detection on retinal fundus images. IPSCL combines position guidance, inexact supervised learning based on a novel Radial Cup–Disc Ratio (RCDR), and sector-level contrastive learning to leverage anatomical correlations and enhance fine-grained feature discrimination. Experiments on the original RNFLD dataset and its extended version show that IPSCL outperforms state-of-the-art methods, achieving sector-level F1 score of 83.84% and image-level F1 score of 94.98%. The method effectively detects subtle and diffuse defects while reducing annotation dependence. This work provides a reliable and interpretable solution for clinical RNFLD detection. Full article

Glaucoma is a leading cause of blindness, and early detection is critical. Convolutional neural networks (CNNs) have shown impressive performance in glaucoma diagnosis, but their black-box nature remains a barrier to clinical use. Existing explainable AI (XAI) methods such as Grad-CAM have limitations in identifying and quantifying subtle regional features. In this study, we propose a method to clarify what CNNs focus on by analyzing how model performance changes under localized adversarial noise. Using VGG16 for glaucoma classification, we applied noise generated by the Fast Gradient Sign Method (FGSM) to the whole fundus image and to specific subregions, then compared the impact on classification performance. Results showed that perturbations to the optic disc, especially its outer margin, had the greatest effect on model performance. This suggests that the CNN captures fine anatomical features such as optic disc cupping and neuroretinal rim thinning, which aligns with what ophthalmologists typically look for. At the same time, perturbations in the macula and perivascular regions also affected performance, indicating gaps between current clinical diagnostic criteria and the CNN’s decision-making process. This approach can help establish the clinical reliability of CNNs and may also reveal features that have not been recognized in conventional clinical practice. Full article

Background: Eosinophilic esophagitis (EoE) is a chronic immune-mediated disease that is increasingly recognized as a systemic inflammatory condition. Its potential association with subsequent coded ocular diagnoses has not been well characterized in large-scale longitudinal studies. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network, which aggregates de-identified electronic health records from multiple international healthcare systems. Adults aged ≥18 years with at least two clinical encounters between 2005 and 2024 were included. Patients with EoE (ICD-10-CM K20.0) were identified as the exposure cohort, while individuals undergoing routine health examinations without EoE served as controls. Those with prior ocular disease, malignancy, or death were excluded. Propensity score matching (1:1) was used to balance demographics, body mass index, comorbidities, medication use, and socioeconomic factors. The primary outcomes were coded ocular diagnostic categories identified using ICD-10-CM codes. To reduce the likelihood of including pre-existing conditions, ocular disease events diagnosed within 3 months after the index date were excluded from the analysis. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated. Sensitivity analyses incorporated alternative exposure definitions, washout periods, and follow-up durations, with additional stratification by age, sex, and race. Results: After matching, 64,613 patients were included in each cohort. EoE diagnostic coding was associated with a higher subsequent occurrence of several coded ocular diagnostic categories, including visual disturbance and blindness (HR = 1.521; 95% CI: 1.383–1.673), disorders of refraction and accommodation (HR = 1.324; 95% CI: 1.188–1.474), lacrimal system disorders (HR = 1.504; 95% CI: 1.274–1.775), cataract (HR = 1.637; 95% CI: 1.384–1.935), glaucoma (HR = 1.463; 95% CI: 1.157–1.849), and disorders of the vitreous body and globe (HR = 1.903; 95% CI: 1.510–2.399). These findings should be interpreted cautiously because several outcomes, such as visual disturbance, disorders of refraction and accommodation, and ocular pain, were broad diagnostic categories and may be susceptible to detection or coding practices. Conclusions: In this large-scale EHR-based cohort study, EoE diagnostic coding was associated with a higher subsequent occurrence of several coded ocular diagnostic categories. These findings should be interpreted as exploratory associations rather than evidence of direct causal or mechanistic relationships, particularly for broad or detection-prone outcomes such as visual disturbance, disorders of refraction and accommodation, and ocular pain. Full article

Background/Objectives: Conjunctival hyperemia is a common clinical finding in clinical practice; however there are significant differences between graders. Vessel detection using deep-learning approaches could enable more objective measures. We aimed to evaluate vascular metrics derived from automated vessel detection and compare these metrics with manual severity gradings. Methods: Slit lamp images from 139 glaucoma patients were included. Images from 103 participants were used as the primary development dataset and the remaining as a validation subset. The images were independently graded by two graders for conjunctival hyperemia using the Efron Grading Scheme. Conjunctival vessels were detected using an automated vessel detection pipeline based on semi-supervised learning. Vessel density, fractal dimension and tortuosity were calculated and compared with the manual Efron grades. Results: Grading of conjunctival hyperemia between the two graders were consistent (Spearman’s rho: 0.79; ICC: 0.79 [95%CI: 0.72–0.84]) but showed significant differences with a higher proportion of differences in the moderate grades. Of the vascular metrics, vessel density showed significant associations with the individual Efron grading and against the mean Efron grading (0.78, p < 0.001). Fractal dimension was significantly associated with the mean Efron grading (0.55, p < 0.001). Agreements were similar in the subset (vessel density, 0.80, p < 0.001; fractal dimension 0.62, p < 0.001). Vessel tortuosity showed lower agreements (<0.23). Conclusions: Vessel density and fractal dimension showed significant associations with manual Efron gradings. These metrics could be potentially used to enable more objective and interpretable measures of conjunctival hyperemia severity. Full article

Purpose: To investigate the short-term impact of scleral lens wear on intraocular pressure (IOP) and retinal nerve fiber layer (RNFL) thickness. We hypothesized that scleral lens wear would produce a measurable elevation in IOP accompanied by detectable RNFL thinning compared with soft contact lens wear. Methods: This prospective, randomized, contralateral-eye crossover study included 31 healthy participants (mean age: 26 ± 3 years). Each participant wore a 16.5 mm scleral lens over one eye and a soft contact lens over the fellow eye for 8 h, with assignments reversed between visits. IOP was measured using two tonometers: a transpalpebral Diaton tonometer and a non-contact tonometer (NCT), and RNFL thickness was measured by optical coherence tomography at four time points: pre- and post-lens application, and pre- and post-lens removal. Results: Eyes fitted with scleral lenses exhibited a significant IOP increase immediately after lens application (pre-application: 11 ± 3 mmHg vs. post-application: 16 ± 4 mmHg, p < 0.001), sustained throughout 8 h of wear (pre-removal: 16 ± 4 mmHg), and returned to baseline after removal (11 ± 3 mmHg). No significant IOP changes were observed in soft contact lens-wearing eyes (p > 0.05). Scleral lens wear was also associated with small but statistically significant peripapillary RNFL thinning (pre-application: 110 ± 11 µm vs. post-application: 107 ± 11 µm, p < 0.001), which returned to baseline after lens removal. No significant RNFL changes were observed with soft contact lens wear (p > 0.05). Bland–Altman analysis revealed poor agreement between Diaton and NCT measurements, consistent with the published literature on transpalpebral tonometry. Conclusions: Short-term scleral lens wear was associated with transient IOP elevation and peripapillary RNFL thinning, both reversible upon lens removal, in healthy young adults. These findings highlight the need for further longitudinal investigation in at-risk populations such as those with ocular hypertension, keratoconus, or early glaucoma before clinical monitoring recommendations can be established. Full article

Glaucoma is a leading cause of irreversible vision loss, and its early diagnosis remains critically important yet challenging. Traditional assessment based on the cup-to-disc ratio is often insufficient at early stages, whereas the retinal vascular network can provide additional quantitative biomarkers. This study develops and validates a binary classification method for distinguishing healthy from glaucomatous fundus images by combining deep-learning-based vessel segmentation, fractal and multifractal analysis, and textural features. The public ORIGA dataset is utilized. Images are converted to grayscale using three alternative approaches, followed by Gray-Level Co-occurrence Matrix texture analysis and fractal analysis based on the differential box-counting method. Vessel segmentation is implemented via a U-Net neural network trained on a combination of public datasets, after which multifractal analysis is performed on the resulting binary masks. The extracted features are used to train and compare several machine learning models with hyperparameter optimization. The best-performing model among ONH-based features (Random Forest) achieves 75.00%; however, a logistic regression model using multifractal parameters and CDR reaches 86.17%, substantially outperforming the CDR-only baseline (66.15%). Notably, while classical fractal dimension shows only marginal differences (1–2% relative change) between groups, multifractal parameters reveal distinct changes: the multifractal spectrum width $\Delta \alpha$ increases markedly and the minimum singularity exponent ${\alpha }_{\min }$ decreases in glaucomatous eyes, indicating increased heterogeneity of the vascular network. These findings suggest that multifractal characteristics of the vascular network can serve as reliable and sensitive biomarkers for automated glaucoma screening, offering clear advantages over classical fractal analysis. Full article