The geometry of retinal layers is an important imaging feature for the diagnosis of some ophthalmic diseases. In recent years, retinal layer segmentation methods for optical coherence tomography (OCT) images have emerged one after another, and huge progress has been achieved. However, challenges due to interference factors such as noise, blurring, fundus effusion, and tissue artifacts remain in existing methods, primarily manifesting as intra-layer false positives and inter-layer boundary deviation. To solve these problems, we propose a method called Tightly combined Cross-Convolution and Transformer with Boundary regression and feature Polarization (TCCT-BP). This method uses a hybrid architecture of CNN and lightweight Transformer to improve the perception of retinal layers. In addition, a feature grouping and sampling method and the corresponding polarization loss function are designed to maximize the differentiation of the feature vectors of different retinal layers, and a boundary regression loss function is devised to constrain the retinal boundary distribution for a better fit to the ground truth. Extensive experiments on four benchmark datasets demonstrate that the proposed method achieves state-of-the-art performance in dealing with problems of false positives and boundary distortion. The proposed method ranked first in the OCT Layer Segmentation task of GOALS challenge held by MICCAI 2022. The source code is available at https://www.github.com/tyb311/TCCT.
Background: To contrast retinal and choroidal vascular alterations between carotid artery dissection (CAD) and carotid artery stenosis (CAS), and analyze their associations with cerebral infarction patterns and stenosis severity. Methods: This observational study included stenotic‐degree‐matched patients with CAD and patients with CAS. Retinal ischemic signs—cotton wool spots, retinal artery occlusion, and paracentral acute middle maculopathy—were evaluated using fovea‐centered optical coherence tomography/angiography. Retinal vascular density and choroidal metrics (choriocapillaris density, thickness, vascular volume, and index) were analyzed. Results: The final analysis included 863 eyes from 467 participants (85 with CAD, 274 with CAS, 108 controls). Patients with CAD showed a significantly higher prevalence of paracentral acute middle maculopathy than Patients with CAS (6.4% versus 0.61%, P <0.001). Although ipsilateral eyes in both groups showed reduced retinal vascular density ( P <0.017), a divergent pattern was observed in the choroid: CAD‐affected eyes demonstrated significantly higher choriocapillaris density, choroidal vascular volume, and choroidal thickness relative to both control and CAS‐affected eyes ( P <0.017). Clinically, retinal ischemic signs in CAD were significantly associated with cerebral infarction ( P =0.022) and severe stenosis/occlusion ( P =0.008). Notably, retinal signs correlated with cerebral infarction topography: subcortical infarctions were predominantly associated with paracentral acute middle maculopathy, whereas cortical infarctions were linked to cotton wool spots and retinal artery occlusion ( P =0.004). Conclusions: Compared with atherosclerotic stenosis, CAD exhibits a distinct oculovascular phenotype characterized by a higher burden of acute retinal ischemia but compensatory choroidal hyperperfusion. Retinal imaging serves as a noninvasive window into CAD pathophysiology, offering biomarkers for assessing hemodynamic stability and thromboembolic risk. Registration: URL: www.chictr.org.cn ; Unique identifier: ChiCTR2300074640.
13 August 2026 in Expert Systems with Applications
Objectives: To estimate the incidence of reported retinal detachment (RD) after primary anti-vascular endothelial growth factor (anti-VEGF) treatment for retinopathy of prematurity (ROP) and to explore study-level factors associated with RD incidence. Methods: PubMed, EMBASE, Cochrane Library, and Web of Science were searched up to 24 May 2026, for English-language studies reporting RD outcomes after anti-VEGF monotherapy for ROP. Pooled incidence was estimated with a random-effects model using Freeman–Tukey double arcsine transformation. Heterogeneity, subgroup differences, publication bias, and leave-one-out sensitivity were assessed. Results: Twenty-seven studies (5598 eyes; 135 RD events) were included, comprising twelve case series, thirteen cohort studies, and two randomized controlled trials; thus, the evidence was predominantly retrospective. The pooled RD incidence was 1.4% (95% CI, 0.5–2.7%; I2 = 80.8%). The estimate remained stable on leave-one-out sensitivity analysis; the exclusion of Tong et al. markedly reduced heterogeneity. Higher RD incidence was observed in Asian studies, single-center studies, and studies with predominantly Zone I ROP or aggressive ROP (AROP). Incidence varied by postmenstrual age at treatment, gestational age, birth weight, and anti-VEGF agent. No significant publication bias was detected. Conclusions: RD after primary anti-VEGF therapy for ROP is uncommon but sight-threatening, with higher incidence associated with greater disease severity and postmenstrual age. Prolonged, risk-stratified follow-up is warranted for infants with Zone I or AROP. Standardized RD definitions, subtype classification, and severity-adjusted prospective comparisons are needed.
20 October 2026 in Journal of the American Heart Association
Background: Severe carotid artery stenosis or occlusion (CASO) is associated with white matter injury and cognitive impairment, but markers of accompanying microvascular dysfunction remain incompletely characterized. We investigated the relationships among retinal microvascular abnormalities on optical coherence tomography angiography, white matter hyperintensities (WMH), and cognitive impairment in severe unilateral CASO. Methods: In this multicenter cross‐sectional study, 860 participants were included: 566 patients with severe unilateral CASO without acute cerebral infarction and 294 control participants with similar age and sex distributions. All participants underwent multimodal imaging, including automated WMH segmentation on brain magnetic resonance imaging and macular optical coherence tomography angiography. Cognitive function was assessed using the Montreal Cognitive Assessment. Results: Compared with controls, patients with CASO showed lower vessel length density across retinal layers and greater WMH burden. Within the group with CASO , the retina ipsilateral to the stenosis showed lower vessel length density and greater vessel dilation, paralleling the greater burden of deep WMH in the ipsilateral hemisphere (all adjusted P <0.05). Retinal microvascular abnormalities and greater WMH burden were both associated with worse concurrent cognitive performance. Adding optical coherence tomography angiography metrics to clinical and WMH ‐based models provided modest additional discrimination for identifying concurrent cognitive impairment in internal cross‐validation, increasing the area under the curve from 0.652 to 0.695 ( Δarea under the curve , 0.043 [95% CI , 0.021–0.064]). Conclusions: In severe unilateral CASO, retinal microvascular abnormalities were associated with ipsilateral white matter injury and lower cognitive performance. Optical coherence tomography angiographymetrics provided modest complementary discriminatory information beyond conventional clinical variables and WMH for identifying concurrent cognitive impairment, although external validation is required. Registration: URL: https://www.chictr.org.cn/ . Unique identifier: ChiCTR2300074640.