Search

High‐fidelity segmentation of blood vessels plays a pivotal role in numerous biomedical applications, such as injection assistance, cancer detection, various surgeries, and vein authentication. Near‐infrared (NIR) transillumination imaging is an effective and safe method to visualize the subcutaneous blood vessel network. However, such images are severely blurred because of the light scattering in body tissues. Inspired by the Vision Transformer model, this paper proposes a novel deep learning network known as transformer connection (TRC)‐Unet to capture global blurred and local clear correlations while using multi‐layer attention. Our method mainly consists of two blocks, thereby aiming to remap skip connection information flow and fuse different domain features. Specifically, the TRC extracts global blurred information from multiple layers and suppresses scattering to increase the clarity of vessel features. Transformer feature fusion eliminates the domain gap between the highly semantic feature maps of the convolutional neural network backbone and the adaptive self‐attention maps of TRCs. Benefiting from the long‐range dependencies of transformers, we achieved competitive results in relation to various competing methods on different data sets, including retinal vessel segmentation, simulated blur image segmentation, and real NIR blood vessel image segmentation. Moreover, our method remarkably improved the segmentation results of simulated blur image data sets and a real NIR vessel image data set. The quantitative results of ablation studies and visualizations are also reported to demonstrate the superiority of the TRC‐Unet design. © 2024 The Author(s). IEEJ Transactions on Electrical and Electronic Engineering published by Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

The study aimed to evaluate the impact of compensating retinal nerve fiber layer (RNFL) thickness for demographic and anatomical factors on glaucoma detection in Chinese and Indian adults. A population‐based study included 1995 healthy participants (1076 Chinese and 919 Indians) to construct a multivariable linear regression compensation model. This model was applied to 357 Chinese glaucoma patients, 357 healthy Chinese, and 357 healthy Indians using Cirrus spectral‐domain optical coherence tomography (OCT). The compensated RNFL thickness considered age, refractive error, optic disc parameters, and retinal vessel density. Results showed that although the average RNFL thickness was significantly higher in Chinese participants compared to Indians, the compensation model reduced this difference to nonsignificance. Moreover, the compensation model significantly improved the area under the receiver operating characteristic curve (0.90 vs. 0.78; p<0.001), sensitivity (75% vs. 51%), and specificity (67% vs. 32%) in distinguishing Chinese glaucoma patients from healthy Indian individuals. The compensation model significantly enhanced the diagnostic accuracy of RNFL thickness in distinguishing glaucoma in the Chinese ethnic group compared to the OCT instrument's default values. These results suggest that modifying RNFL measurements based on individual characteristics can yield substantial benefits for glaucoma detection across ethnicities. [ABSTRACT FROM AUTHOR]

Objectives: The effects of HIV and antiretroviral therapy (ART) on microvascular function are poorly explored. We compared retinal vessel functional responses to flicker light‐induced provocation (FLIP) in people living with HIV (PLWH) and people living without HIV (PLWoutH). Methods: We included 115 PLWH and 51 PLWoutH with a median age of 41 years. Treated PLWH received similar first‐line fixed‐dose combination ART. Clinical characteristics and retinal vessels functional responses to FLIP were compared in (a) PLWH and PLWoutH; and (b) PLWH groups stratified by the median of (i) CD4‐count (511 cells/mm3), (ii) viral load (50 copies/mL), and (iii) ART duration (57.6 months). Results: PLWH were older, smoked more, and had a lower prevalence of hypertension than PLWoutH (p < 0.05). Almost 64% of PLWH were infected for more than 5 years. Retinal vessel responses to FLIP were similar between PLWH and PLWoutH after taking confounders into account. In addition, PLWH subgroups stratified according to immuno‐virological status by CD4‐count, viral load, and ART duration showed no differences in retinal vessel responses to FLIP. Conclusion: Living with HIV and receiving ART were not associated with altered microvascular function as assessed with dynamic retinal vessel analysis in a South African case–control study. [ABSTRACT FROM AUTHOR]

ABSTRACT Our study represents the first attempts to describe the ultrastructure features in addition to SEM‐EDX analysis of the pecten oculi of the nocturnal, visually active Eurasian stone‐curlew. The present study was carried out on 10 normal pecten oculi from the migrant Eurasian stone‐curlew (Burhinus oedicnemus) that was captured from the Northern Egyptian coast, Edku Lake, Edku, Behera, Egypt. The intraocular folded quadrilateral fan black pecten oculi was observed on the posteroinferior wall of the eyeball. The pecten oculi had 18–20 pleats, with narrow spaces in‐between. Our SEM findings showed a tortuous, coiled pecteneal surface due to numerous blood capillaries. The apical part was oriented toward the ciliary body, but the basal part was derived from the retina, just behind the optic nerve. The outer pleat surfaces are covered by the vitreopecteneal limiting membrane, which carries numerous hyalocyte cells on the outer pleat surface. The transverse sections revealed numerous blood capillaries surrounded by numerous melanin granules. Each basal part was surrounded by a thin, elevated ridge. Our SEM‐EDX analysis revealed that carbon is the highest (half percent), oxygen represents about one‐third, and nitrogen is one‐fifth percent of all elements. The lowest elements are sulfate at the apical part and phosphorus in the middle and basal parts. In conclusion, our findings reveal that the number of pecteneal pleats is related to the nocturnal active visual pattern and the lifestyle of this migratory bird. [ABSTRACT FROM AUTHOR]

Retinopathy of prematurity (ROP) is a severe retinal disease in premature infants characterized by pathological neovascularization, obliteration of retinal vessels and increased vessel tortuosity. Currently, there are no completely satisfactory treatments for ROP. Pigment epithelium‐derived factor (PEDF), a potent inhibitor of angiogenesis, appears late in gestation and its deficiency may be linked to development of ROP. This study investigates the preclinical efficacy of PEDF protein alone or in combination with VEGF antagonists for treating ROP. The safety of PEDF protein in the rat eye was assessed using functional in vivo measurements and histology. The efficacy of intravitreal injections (IVI) of various treatments was evaluated in a rat oxygen‐induced retinopathy (OIR) model using in vivo imaging and flatmount analyses. No functional or histological side‐effects were found in rat eyes after intravitreal PEDF protein injection. PEDF protein alone or combined with anti‐VEGF drugs significantly reduced pathological neovascularization and vessel obliteration, comparable to the effects of anti‐VEGF drugs alone. Regarding arterial tortuosity, treatment with a combination of PEDF, and VEGF antagonist was more effective than treatment with anti‐VEGF alone. IVI of PEDF protein is safe. PEDF protein alone or combined with VEGF antagonists shows similar efficacy in reducing pathological neovascularization and vessel obliteration as anti‐VEGF agents. Furthermore, only treatments involving PEDF protein, alone or with VEGF antagonists, significantly improved the quality of retinal vasculature. Thus, PEDF protein alone or combined with anti‐VEGF agents presents a promising alternative to current anti‐VEGF treatments for ROP. [ABSTRACT FROM AUTHOR]

Automatic segmentation of the fundus retinal vessels and accurate classification of the arterial and venous vessels play an important role in clinical diagnosis. This article proposes a fundus retinal vascular segmentation and arteriovenous classification network that combines the adversarial training and attention mechanism to address the issues of fundus retinal arteriovenous classification error and ambiguous segmentation of fine blood vessels. It consists of three core components: discriminator, generator, and segmenter. In order to address the domain shift issue, U‐Net is employed as a discriminator, and data samples for arterial and venous vessels are generated with a generator using an unsupervised domain adaption (UDA) approach. The classification of retinal arterial and venous vessels (A/V) as well as the segmentation of fine vessels is improved by adding a self‐attention mechanism to improve attention to vessel edge features and the terminal fine vessels. Non‐strided convolution and non‐pooled downsampling methods are also used to avoid losing fine‐grained information and learning less effective feature representations. The performance of multi‐class blood vessel segmentation is as follows, per test results on the DRIVE dataset: F1‐score (F1) has a value of 0.7496 and an accuracy of 0.9820. The accuracy of A/V categorization has increased by 1.35% when compared to AU‐Net. The outcomes demonstrate that by enhancing the baseline U‐Net, the strategy we suggested enhances the automated classification and segmentation of blood vessels. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the association between pretreatment blood flow velocity in the choroid and optic nerve head (ONH) and retinal oxygen metabolism in the acute uveitic phase and the development of 'sunset glow fundus' in Vogt‐Koyanagi‐Harada (VKH) disease. Methods: Retrospective analysis of 41 patients (82 eyes). Laser speckle flowgraphy and retinal oximetry measurements were performed at the presentation. The main outcome measure was the development of 'sunset glow fundus'. Results: Twenty patients (40 eyes) presented in the phase preceding anterior segment inflammation (early presentation), and 21 patients (42 eyes) presented with anterior segment inflammation (late presentation). In ONH, mean blur rate (MBR)‐vessel, representing blood flow velocity in retinal vessels, was significantly lower in the late presentation group, while choroidal MBR was not significantly different. The late presentation group had significantly lower oxygen saturation in retinal venules, a higher arteriovenous oxygen saturation difference and a smaller calibre of retinal arterioles compared with the early presentation group. Eyes that subsequently developed 'sunset glow fundus' had significantly lower ONH MBR‐vessels, lower oxygen saturation in retinal venules, a higher arteriovenous oxygen saturation difference and a smaller calibre of retinal arterioles compared with eyes without 'sunset glow fundus'. ONH MBR‐vessel had a significant negative correlation with arteriovenous oxygen saturation difference and a significant positive correlation with calibre of retinal arterioles. Conclusions: In the acute uveitic phase of VKH disease, the development of 'sunset glow fundus' is associated with pretreatment reduced retinal blood flow velocity, calibre of retinal arterioles and oxygen saturation in retinal venules, as well as an increased arteriovenous oxygen saturation difference. [ABSTRACT FROM AUTHOR]

Mesoscopic photoacoustic imaging (PAI) enables label‐free visualization of vascular networks in tissues with high contrast and resolution. Segmenting these networks from 3D PAI data and interpreting their physiological and pathological significance is crucial yet challenging due to the time‐consuming and error‐prone nature of current methods. Deep learning offers a potential solution; however, supervised analysis frameworks typically require human‐annotated ground‐truth labels. To address this, an unsupervised image‐to‐image translation deep learning model is introduced, the Vessel Segmentation Generative Adversarial Network (VAN‐GAN). VAN‐GAN integrates synthetic blood vessel networks that closely resemble real‐life anatomy into its training process and learns to replicate the underlying physics of the PAI system in order to learn how to segment vasculature from 3D photoacoustic images. Applied to a diverse range of in silico, in vitro, and in vivo data, including patient‐derived breast cancer xenograft models and 3D clinical angiograms, VAN‐GAN demonstrates its capability to facilitate accurate and unbiased segmentation of 3D vascular networks. By leveraging synthetic data, VAN‐GAN reduces the reliance on manual labeling, thus lowering the barrier to entry for high‐quality blood vessel segmentation (F1 score: VAN‐GAN vs. U‐Net = 0.84 vs. 0.87) and enhancing preclinical and clinical research into vascular structure and function. [ABSTRACT FROM AUTHOR]

Cancer is one of the leading causes of death in the world, with radiotherapy as one of the treatment options. Radiotherapy planning starts with delineating the affected area from healthy organs, called organs at risk (OAR). A new approach to automatic OAR segmentation in the chest cavity in Computed Tomography (CT) images is presented. The proposed approach is based on the modified U‐Net architecture with the ResNet‐34 encoder, which is the baseline adopted in this work. The new two‐branch CS‐SA U‐Net architecture is proposed, which consists of two parallel U‐Net models in which self‐attention blocks with cosine similarity as query‐key similarity function (CS‐SA) blocks are inserted between the encoder and decoder, which enabled the use of consistency regularisation. The proposed solution demonstrates state‐of‐the‐art performance for the problem of OAR segmentation in CT images on the publicly available SegTHOR benchmark dataset in terms of a Dice coefficient (oesophagus—0.8714, heart—0.9516, trachea—0.9286, aorta—0.9510) and Hausdorff distance (oesophagus—0.2541, heart—0.1514, trachea—0.1722, aorta—0.1114) and significantly outperforms the baseline. The current approach is demonstrated to be viable for improving the quality of OAR segmentation for radiotherapy planning. [ABSTRACT FROM AUTHOR]

Background: CT‐image segmentation for liver and hepatic vessels can facilitate liver surgical planning. However, time‐consuming process and inter‐observer variations of manual segmentation have limited wider application in clinical practice. Purpose: Our study aimed to propose an automated deep learning (DL) segmentation algorithm for liver and hepatic vessels on portal venous phase CT images. Methods: This retrospective study was performed to develop a coarse‐to‐fine DL‐based algorithm that was trained, validated, and tested using private 413, 52, and 50 portal venous phase CT images, respectively. Additionally, the performance of the DL algorithm was extensively evaluated and compared with manual segmentation using an independent clinical dataset of preoperative contrast‐enhanced CT images from 44 patients with hepatic focal lesions. The accuracy of DL‐based segmentation was quantitatively evaluated using the Dice Similarity Coefficient (DSC) and complementary metrics [Normalized Surface Dice (NSD) and Hausdorff distance_95 (HD95) for liver segmentation, Recall and Precision for hepatic vessel segmentation]. The processing time for DL and manual segmentation was also compared. Results: Our DL algorithm achieved accurate liver segmentation with DSC of 0.98, NSD of 0.92, and HD95 of 1.52 mm. DL‐segmentation of hepatic veins, portal veins, and inferior vena cava attained DSC of 0.86, 0.89, and 0.94, respectively. Compared with the manual approach, the DL algorithm significantly outperformed with better segmentation results for both liver and hepatic vessels, with higher accuracy of liver and hepatic vessel segmentation (all p < 0.001) in independent 44 clinical data. In addition, the DL method significantly reduced the manual processing time of clinical postprocessing (p < 0.001). Conclusions: The proposed DL algorithm potentially enabled accurate and rapid segmentation for liver and hepatic vessels using portal venous phase contrast CT images. [ABSTRACT FROM AUTHOR]

Purpose Design Methods Results Conclusion To characterise the retinal vasculometry of a Danish eye and vision cohort and examine associations with systolic blood pressure (BP), diastolic BP, mean arterial BP, and intraocular pressure (IOP).Longitudinal study.The retinal vasculature of fundus images from the FOREVER (Finding Ophthalmic Risks and Evaluating the Value of Eye exams and their predictive Reliability) cohort was analysed using a fully automated image analysis program. Longitudinal associations of retinal vessel morphology at follow‐up visit with IOP (baseline and follow‐up) and BP (follow‐up) were examined using multilevel linear regression models adjusting for age, sex and retinal vasculometry at baseline as fixed effects and person as random effect. Width measurements were additionally adjusted for the spherical equivalent.A total of 2089 subjects (62% female) with a mean age of 61 (standard deviation 8) years and a mean follow‐up period of 4.1 years (SD 0.6 years) were included. The mean arteriolar diameter was approximately 20% thinner than the mean venular diameter, and venules were about 21%–23% less tortuous than arterioles. BP at follow‐up was associated with decreased arteriolar diameter from baseline to follow‐up. After adjusting for baseline IOP, IOP at follow‐up was associated with increased arteriolar tortuosity above baseline (0.59%, 95% CI 0.08–1.10, p‐value 0.024).In a Danish eye and vision cohort, variations in BP and alterations in IOP over time were associated with changes in the width and tortuosity of retinal vessels. Our findings contribute novel insights into retinal vascular alterations over time. [ABSTRACT FROM AUTHOR]

Accurate segmentation of retinal vessels is crucial for the early diagnosis and treatment of eye diseases, for example, diabetic retinopathy, glaucoma, and macular degeneration. Due to the intricate structure of retinal vessels, it is essential to extract their features with precision for the semantic segmentation of medical images. In this study, an improved deep learning neural network was developed with a focus on feature extraction based on the U‐Net structure. The enhanced U‐Net combines the architecture of convolutional neural networks (CNNs) with SE blocks (squeeze‐and‐excitation blocks) to adaptively extract image features after each U‐Net encoder's convolution. This approach aids in suppressing nonvascular regions and highlighting features for specific segmentation tasks. The proposed method was trained and tested on the DRIVECHASE_DB1 and STARE datasets. As a result, the proposed model had an algorithmic accuracy, sensitivity, specificity, Dice coefficient (Dc), and Matthews correlation coefficient (MCC) of 95.62/0.9853/0.9652, 0.7751/0.7976/0.7773, 0.9832/0.8567/0.9865, 82.53/87.23/83.42, and 0.7823/0.7987/0.8345, respectively, outperforming previous methods, including UNet++, attention U‐Net, and ResUNet. The experimental results demonstrated that the proposed method improved the retinal vessel segmentation performance. [ABSTRACT FROM AUTHOR]

The objective of this study was to investigate the effects and molecular mechanisms of tetrahedral framework nucleic acids‐microRNA22 (tFNAs‐miR22) on inhibiting pathological retinal neovascularization (RNV) and restoring physiological retinal vessels. A novel DNA nanocomplex (tFNAs‐miR22) was synthesised by modifying microRNA‐22 (miR22) through attachment onto tetrahedral frame nucleic acids (tFNAs), which possess diverse biological functions. Cell proliferation, wound healing, and tube formation were employed for in vitro assays to investigate the angiogenic function of cells. Oxygen‐induced retinopathy (OIR) model was utilised to examine the effects of reducing pathological neovascularization (RNV) and inhibiting vascular occlusion in vivo. In vitro, tFNAs‐miR22 demonstrated the ability to penetrate endothelial cells and effectively suppress cell proliferation, tube formation, and migration in a hypoxic environment. In vivo, tFNAs‐miR22 exhibited promising results in reducing RNV and promoting the restoration of normal retinal blood vessels in OIR model through modulation of the Wnt pathway. This study provided a theoretical basis for the further understanding of RNV, and highlighted the innovative and potential of tFNAs‐miR22 as a therapeutic option for ischemic retinal diseases. [ABSTRACT FROM AUTHOR]

This article presents a study on a new genetic rearrangement that is linked to a condition called PRPF31-associated retinitis pigmentosa (RP). RP is a genetic eye disorder that causes the degeneration of certain cells in the retina, leading to vision problems. The researchers discovered a specific deletion on chromosome 19 that includes the PRPF31 gene, which is associated with RP. They also observed reduced levels of PRPF31 in RP patients. This research offers valuable insights into the genetic causes of RP and could potentially contribute to the development of targeted treatments. Additionally, the study explores a specific mechanism called Alu/Alu-mediated rearrangement (AAMR) that is responsible for this genetic rearrangement. [Extracted from the article]

Purpose: To evaluate if retinal vascular calibers and systemic risk factors in patients with no or minimal diabetic retinopathy (DR) can predict risk of long‐term progression to proliferative diabetic retinopathy (PDR). Methods: This was a matched case–control study of patients with diabetes having no or minimal DR at baseline with (cases) or without (controls) subsequent development of PDR. We collected six‐field, 45‐degree retinal images, demographic and clinical data from the Funen Diabetes Database. Results: We included 52 eyes from 39 cases and 107 eyes from 89 controls matched on sex, age, type of diabetes, time from first to last screening episode and baseline DR level. Cases had higher HbA1c (73 vs. 55 mmoL/moL; p < 0.001), triglycerides (1.32 vs. 1.16 mmoL/L; p = 0.02) and longer duration of diabetes (19 vs. 14 years; p = 0.01), but the groups did not differ in calibers of retinal arterioles (229 vs. 227 μm; p = 0.49), venules (289 vs. 290 μm; p = 0.83) or the arterio‐to‐venule ratio (0.78 vs. 0.77; p = 0.86).In a multivariable logistic regression model with cluster robust standard error, HbA1c (OR 1.54 per 10 mmoL/moL; 95%‐CI: 1.15–2.07; p = 0.004), triglyceride (OR 1.39 per 1 mmoL/L; 95%‐CI: 1.03–1.86; p = 0.03) and duration of diabetes (OR 1.09 per year; 95%‐CI: 1.03–1.16; p = 0.003) were independent risk factors for PDR. Conclusion: Retinal vascular calibers did not predict long‐term development of PDR in contrast to well‐established risk factors like HbA1c, triglyceride and duration of diabetes. [ABSTRACT FROM AUTHOR]

Segmentation is crucial in diagnosing retinal diseases by accurately identifiying retinal vessels. This paper addresses the complexity of segmenting retinal vessels, highlighting the need for precise analysis of blood vessel structures. Despite the progress made by convolutional neural networsks (CNNs) in image segmentation, their limitations in capturing the global structure of retinal vsessels and maintaining segmentation continuity present challenges. To tackle these issues, our proposed network integrates graph convolutional networks (GCNs) and attention mechansims. This allows the model to consider pixel relationships and learn vessel graphical structures, significantly improving segmentation accuracy. Additionally, the attentional feature fusion module, including pixel‐wise and channel‐wise attention mechansims within the U‐Net architecture, refines the model's focus on relevant features. This paper emphasizes the importance of continuty preservation, ensuring an accurate representation of pixel‐level information and structural details during sefmentation. Therefore, our method performs as an effective solution to overcome challenges in retinal vessel segmentation. The proposed method outperformed the state‐of‐the‐art approaches on DRIVE (Digital Retinal Images for Vessel Extraction) and STARE (Structed Analysis of the Retina) datasets with accuracies of 0.12% and 0.14%, respecttively. Importantly, our proposed approach excelled in delineating slender and diminutive blood vessels, crucial for diagnosing vascular‐related diseases. Implementation is accessible on https://github.com/CVLab‐SHUT/VGA‐Net. [ABSTRACT FROM AUTHOR]

Click‐through rate (CTR) prediction is a pivotal challenge in recommendation systems. Existing models are prone to disturbances from noise and redundant features, hindering their ability to fully capture implicit and higher‐order feature interactions present in sparse feature data. Moreover, conventional dual‐tower models overlook the significance of layer‐level feature interactions. To address these limitations, this article introduces Gate‐enhanced Multi‐space Interactive Neural Networks (GMINN), a novel model for CTR prediction. GMINN adopts a dual‐tower architecture in which a multi‐space interaction layer is introduced after each layer in the dual‐tower deep neural network. This layer allocates features into multiple subspaces and employs matrix multiplication to establish layer‐level interactions between the dual towers. Simultaneously, a field‐aware gate mechanism is proposed to extract crucial latent information from the original features. Experimental validation on publicly available datasets, Criteo and Avazu, demonstrates the superiority of the proposed GMINN model. Comparative analyses against baseline models reveal that GMINN substantially improves up to 4.09% in AUC and a maximum reduction of 7.21% in Logloss. Additionally, ablation experiments provide further validation of the effectiveness of GMINN. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the changes in retinal arterial architecture after treatment with voretigene neparvovec in patients with retinal dystrophy caused by bi‐allelic mutations in the RPE65 gene. Methods: Sixteen eyes treated with voretigene neparvovec at the University Eye Clinic in Tuebingen, Germany, underwent adaptive optics ophthalmoscopy (AO) imaging at baseline and 2 weeks, 1, 3, 6 and 12 months after treatment. Follow‐up was performed in six eyes of four patients. For each eye, five different positions at arterial vessels were selected and the wall‐to‐lumen ratio (WLR), the lumen diameter (LD) and the wall cross‐sectional area (WCSA) were measured by the manufacturer's software over the observational period. Results: Vast retinal atrophy dominated all gained AO images. WLR fluctuated in the observation period without statistically significant change. LD and WCSA changed significantly after 2 weeks from the baseline examination and returned to values similar to baseline thereafter. There were no signs of inflammation such as macrophages or perivascular accumulated fluid visible. Conclusion: AO imaging of the retinal vessels in RPE65‐associated retinal dystrophies (IRD) is challenging. There was no change in the retinal arterial vasculature over the observation period of 12 months that would indicate inflammatory changes. Decrease of the LD and WCSA shortly after treatment might be caused by the perioperative prednisolone intake. AO of retinal vessels can be used as a diagnostic module to complement monitoring the disease and effects of genetic treatments if the acquisition is possible in selected cases. [ABSTRACT FROM AUTHOR]

Purpose: The macular ganglion cell layer (mGCL) is a strong potential biomarker of axonal degeneration in multiple sclerosis (MS). For this reason, this study aims to develop a computer‐aided method to facilitate diagnosis and prognosis in MS. Methods: This paper combines a cross‐sectional study of 72 MS patients and 30 healthy control subjects for diagnosis and a 10‐year longitudinal study of the same MS patients for the prediction of disability progression, during which the mGCL was measured using optical coherence tomography (OCT). Deep neural networks were used as an automatic classifier. Results: For MS diagnosis, greatest accuracy (90.3%) was achieved using 17 features as inputs. The neural network architecture comprised the input layer, two hidden layers and the output layer with softmax activation. For the prediction of disability progression 8 years later, accuracy of 81.9% was achieved with a neural network comprising two hidden layers and 400 epochs. Conclusion: We present evidence that by applying deep learning techniques to clinical and mGCL thickness data it is possible to identify MS and predict the course of the disease. This approach potentially constitutes a non‐invasive, low‐cost, easy‐to‐implement and effective method. [ABSTRACT FROM AUTHOR]

Objective: We aimed to characterize several aspects of retinal vascular dynamics in a patient with arrythmia in order to elicit additional diagnostic information on microvascular dysfunction. Methods: A 68‐year‐old male patient with arrythmia and an age‐ and gender‐matched control subject underwent ocular examination including dynamic retinal vessel assessment with flicker light provocation. Retinal vessel diameters were measured continuously following a standard protocol (IMEDOS Systems, Jena, Germany). The data were evaluated using methods of signal analysis. Results: Retinal vessel response following flicker provocation as well as local structural and functional behavior of retinal vessels were comparable between both individuals. The arrhythmia case demonstrated irregular arterial and venous heart rate (HR) pulsation with an average frequency of 1 Hz. Moreover, the case showed a higher magnitude and larger periods of low‐frequency retinal vessel oscillations as well as lower periodicity of both HR pulsations and low‐frequency vasomotions. Conclusions: Besides numerical examination of irregular HR pulsations in case of arrhythmia, from the direct noninvasive assessment of retinal vessel dynamics one can derive more detailed information on microvascular function including the whole spectrum of retinal arterial and venous pulsations and vasomotions. This may have implications for health screening not limited to atrial fibrillation. [ABSTRACT FROM AUTHOR]

Aim: The retina and brain share similar anatomical and physiological features. Thus, retinal imaging by optical coherence tomography angiography (OCTA) might be a potential tool for the early diagnosis of diabetic cerebral small vessel disease (CSVD). In this study, we aimed to evaluate retinal vascular density (VD) in diabetic CSVD by OCTA imaging and explore the associations between retinal VD and cerebral magnetic resonance imaging (MRI) markers and cognitive function. Methods: In total, 131 patients were enrolled, including CSVD (n = 43) and non‐CSVD groups (n = 88). The VD and foveal avascular zone of the retinal capillary plexus were measured with OCTA. A brain MRI was performed. Results: MRI imaging showed that in the diabetic CSVD group, white matter hyperintensities (WMHs), particularly deep WMHs (58.82%), are the most common MRI marker, followed by cerebral microbleeds in the subtentorial and cortical areas (34.78%). The CSVD group showed increases in the prevalence of cognitive dysfunction (p =.034) and depression (p =.033) and decreases in visuospatial/executive ability and delayed recall ability. In the CSVD group, VDs of the macular superficial vascular plexus (32.93 ± 7.15% vs. 36.97 ± 6.59%, p =.002), intermediate capillary plexus (20.87 ± 4.30% vs. 23.08 ± 4.30%, p =.005) and deep capillary plexus (23.54 ± 5.00% vs. 26.05 ± 4.20%, p =.003) were lower than those of the non‐CSVD group. Multiple linear regression analysis showed that VD of the macular superficial vascular plexus was independently associated with cerebral microbleeds. Meanwhile, VD of the macular intermediate capillary plexus was associated with white matter lacunar infarcts after adjustment. Conclusions: Diabetic CSVDs are characterized by MRI markers, including deep WMHs and cerebral microbleeds, and showed impaired cognition with decreased visuospatial/executive ability and delayed recall ability. OCTA imaging revealed a significant decrease in retinal microvascular perfusion in diabetic CSVD, which was related to MRI markers and cognitive function. OCTA might be a valuable potential measurement for the early diagnosis of CSVD. [ABSTRACT FROM AUTHOR]

Automated partitioning of retinal vessels depicted in fundus images is beneficial in the detection of specific ailments like hypertension and diabetes. However, retinal vessel images have the problems of a large semantic range, more spatial detail, and limited differentiation among the blood vessels and surroundings, which make vessel segmentation challenging. To overcome these obstacles, we designed a new U‐shaped network named SMP‐Net. First, we propose the sequencer‐convolution (SC) module to obtain the ability to extract both local and global features, thereby improving segmentation accuracy. The SC module is used to filter out shallow noise and enable the fusion of deep and shallow features in the maximum skip connection of the U‐shaped network. Then, the residual multi‐kernel pooling (MP) module is designed to obtain additional contextual information while also mitigating the loss of spatial information caused by constant pooling and convolution to improve vessel coherence. Finally, the pixel attention (PA) module redistributes the weight of each pixel using an element‐wise product multiplication operation. This increases the weight of the vascular feature pixels and improves the ability to identify blood vessels in blurred backgrounds. The proposed method has been demonstrated to be effective through sufficient experimentation on publicly available retinal datasets such as DRIVE, STARE, and CHASE_DB1. [ABSTRACT FROM AUTHOR]

Abnormal changes in the retinal vessel width usually indicate the presence of certain diseases. This paper presents a novel model to measure the width of retinal vessels. This method firstly employs cubic spline fitting to determine the vessel cross‐sectional direction at the center point of a vessel profile, then uses the extended edge‐searching mode to detect vessel edge points along the vessel cross‐sectional direction, and finally obtains the width value by calculating the distance between the edge point and the center point. The method can accurately locate vessel edge points and successfully solve the problem of central light reflex in blood vessels. Experimental analyses on a publicly available database show that the proposed approach has excellent performance in accuracy with other existing measurement techniques. It provides an effective idea for analyzing morphological changes in blood vessels and can be extended to obtain the shape features of other objects. [ABSTRACT FROM AUTHOR]

Purpose: Glaucoma leads to pathological loss of axons in the retinal nerve fibre layer at the optic nerve head (ONH). This study aimed to develop a strategy for the estimation of the cross‐sectional area of the axons in the ONH. Furthermore, improving the estimation of the thickness of the nerve fibre layer, as compared to a method previously published by us. Methods: In the 3D‐OCT image of the ONH, the central limit of the pigment epithelium and the inner limit of the retina, respectively, were identified with deep learning algorithms. The minimal distance was estimated at equidistant angles around the circumference of the ONH. The cross‐sectional area was estimated by the computational algorithm. The computational algorithm was applied on 16 non‐glaucomatous subjects. Results: The mean cross‐sectional area of the waist of the nerve fibre layer in the ONH was 1.97 ± 0.19 mm2. The mean difference in minimal thickness of the waist of the nerve fibre layer between our previous and the current strategies was estimated as CIμ (0.95) 0 ± 1 μm (d.f. = 15). Conclusions: The developed algorithm demonstrated an undulating cross‐sectional area of the nerve fibre layer at the ONH. Compared to studies using radial scans, our algorithm resulted in slightly higher values for cross‐sectional area, taking the undulations of the nerve fibre layer at the ONH into account. The new algorithm for estimation of the thickness of the waist of the nerve fibre layer in the ONH yielded estimates of the same order as our previous algorithm. [ABSTRACT FROM AUTHOR]

Segmentation of retinal fragments like blood vessels, optic disc (OD), and optic cup (OC) enables the early detection of different retinal pathologies like diabetic retinopathy (DR), glaucoma, etc. This article proposed a novel deep learning architecture termed as Shell‐Net for the accurate segmentation of the retinal fragments. The main novelty of the architecture relies on the intellectual fusion of two different styled networks for attaining better segmentation results. The lower part of the Shell‐Net (feature condenser) follows the down‐sampling and up‐sampling style, whereas the upper part of the network (feature amplifier) follows the up‐sampling and down‐sampling style of architecture. In addition to this, an additional residual module (feature stabilizer) is integrated with the network to achieve more spatial information from lower levels. The lower part of the network reduces the data through summarization, enabling much scope for precise extraction of heavy details such as thick vessels, OD, and OC. On the contrary, the upper part of the network augments the data using duplication, assisting in the enlargement of minuscule details such as the tiny vessels and boundaries. Experiments were performed on publicly available datasets like Digital Retinal Images for Vessel Extraction (DRIVE), Child Heart and Health Study in England (CHASE_DB1), Structured Analysis of the Retina (STARE), Online Retinal Fundus Image Dataset for Glaucoma Analysis and Research (ORIGA), DRISHTI‐GS1, and Retinal Image database for Optic Nerve Evaluation (RIMONE r1). The network accomplished an average accuracy (ACC) and specificity (SPE) of 0.96 and 0.98 when tested on DRIVE, CHASE_DB1, and STARE datasets respectively for vessel segmentation. Furthermore, it outperformed previously existing models in OD and OC segmentation by achieving an average accuracy of 0.98 with a specificity of 0.99 on the DRISHTI_GS1 dataset. [ABSTRACT FROM AUTHOR]

Diabetes can cause damage to the retina's blood vessels in the eye leading to diabetic retinopathy (DR). The images captured using a fundus camera are used to segment and study the blood vessel damage. Once the retina is damaged, it cannot be repaired. Therefore, early detection of blood vessel damage is the only way to control the progression of the disease. It allows physicians to provide timely and appropriate treatment to patients. The ophthalmologist can manually recognize and mark these vessels based on some clinical and geometrical features; however, it is time‐consuming. Extraction and segmentation play a significant role in showing the difference between healthy and newly developed abnormal vessels. Due to the increased diabetes population, automated systems have been designed to detect retinal blood vessels and assist ophthalmologists. Identifying, extracting, and examining blood vessels are intricate processes, specifically identifying new vessels in the retina. In medical image analysis, artificial intelligence and deep learning techniques have become widely used practices for automatic retinal blood vessel segmentation. We reviewed articles from 1989 to 2023, including handcrafted segmentation to recent deep‐learning techniques with available public datasets. We have concluded this article with an overview of observed parameters, calculations, and future directions for the analysis of retinal images. We believe this review article will help researchers identify research gaps in the field of DR. [ABSTRACT FROM AUTHOR]

This study aimed to examine the impact of diabetes and hypertension on retinal nerve fiber layer (RNFL) thickness components. Optical coherence tomography (OCT) measurements do not consider blood vessel contribution, which this study addressed. We hypothesized that diabetes and/or hypertension would lead to thinner RNFL versus controls due to the vascular component. OCT angiography was used to measure the RNFL in 121 controls, 50 diabetes patients, 371 hypertension patients, and 177 diabetes patients with hypertension. A novel technique separated the RNFL thickness into original (vascular component) and corrected (no vascular component) measurements. Diabetes‐only (98 ± 1.7 µm; p = 0.002) and diabetes with hypertension (99 ± 0.8 µm; p = 0.001) patients had thinner original RNFL versus controls (102 ± 0.8 µm). No difference was seen between hypertension‐only patients (101 ± 0.5 µm; p = 0.083) and controls. After removing the blood vessel component, diabetes/hypertension groups had thinner corrected RNFL versus controls (p = 0.024). Discrepancies in diabetes/hypertension patients were due to thicker retinal blood vessels within the RNFL thickness (p = 0.002). Our findings suggest that diabetes and/or hypertension independently contribute to neurodegenerative thinning of the RNFL, even in the absence of retinopathy. The differentiation of neuronal and vascular components in RNFL thickness measurements provided by the novel technique highlights the importance of considering vascular changes in individuals with these conditions. [ABSTRACT FROM AUTHOR]

Machine vision systems that capture images for visual inspection and recognition tasks must be able to perceive, memorize, and compute any color scene. To achieve this, most of the current visual systems use circuits and algorithms which may reduce efficiency and increase complexity. Herein, a 2D semiconductor tungsten diselenide (WSe2)‐based phototransistor that successfully demonstrates an artificial vision system integrating the processing capability of visual information sensing memory, is reported. Furthermore, based on a 6 × 6 fabricated retinal perception array, artificial visual information sensing memory and processing system are proposed to perform image recognition tasks, which can avoid the time delay and energy consumption caused by data conversion and movement. On the other hand, highly linear symmetric synaptic plasticity can be achieved based on the modulation of carrier types in WSe2 transistors with different thicknesses, facilitating the high level of training and inference accuracy for artificial neural networks. Last, through training and inference simulations, the feasibility of the hybrid synapses for optical neural networks (ONN) is demonstrated. [ABSTRACT FROM AUTHOR]

Background: Bright‐blood lumen and black‐blood vessel wall imaging are required for the comprehensive assessment of aortic disease. These images are usually acquired separately, resulting in long examinations and potential misregistration between images. Purpose: To characterize the performance of an accelerated and respiratory motion‐compensated three‐dimensional (3D) cardiac MRI technique for simultaneous contrast‐free aortic lumen and vessel wall imaging with an interleaved T2 and inversion recovery prepared sequence (iT2Prep‐BOOST). Study Type: Prospective. Population: A total of 30 consecutive patients with aortopathy referred for a clinically indicated cardiac MRI examination (9 females, mean age ± standard deviation: 32 ± 12 years). Field Strength/Sequence: 1.5‐T; bright‐blood MR angiography (diaphragmatic navigator‐gated T2‐prepared 3D balanced steady‐state free precession [bSSFP], T2Prep‐bSSFP), breath‐held black‐blood two‐dimensional (2D) half acquisition single‐shot turbo spin echo (HASTE), and 3D bSSFP iT2Prep‐BOOST. Assessment: iT2Prep‐BOOST bright‐blood images were compared to T2prep‐bSSFP images in terms of aortic vessel dimensions, lumen‐to‐myocardium contrast ratio (CR), and image quality (diagnostic confidence, vessel sharpness and presence of artifacts, assessed by three cardiologists on a 4‐point scale, 1: nondiagnostic to 4: excellent). The iT2Prep‐BOOST black‐blood images were compared to 2D HASTE images for quantification of wall thickness. A visual comparison between computed tomography (CT) and iT2Prep‐BOOST was performed in a patient with chronic aortic dissection. Statistical Tests: Paired t‐tests, Wilcoxon signed‐rank tests, intraclass correlation coefficient (ICC), Bland–Altman analysis. A P value < 0.05 was considered statistically significant. Results: Bright‐blood iT2Prep‐BOOST resulted in significantly improved image quality (mean ± standard deviation 3.8 ± 0.5 vs. 3.3 ± 0.8) and CR (2.9 ± 0.8 vs. 1.8 ± 0.5) compared with T2Prep‐bSSFP, with a shorter scan time (7.8 ± 1.7 minutes vs. 12.9 ± 3.4 minutes) while providing a complementary 3D black‐blood image. Aortic lumen diameter and vessel wall thickness measurements in bright‐blood and black‐blood images were in good agreement with T2Prep‐bSSFP and HASTE images (<0.02 cm and <0.005 cm bias, respectively) and good intrareader (ICC > 0.96) and interreader (ICC > 0.94) agreement was observed for all measurements. Data Conclusion: iT2Prep‐BOOST might enable time‐efficient simultaneous bright‐ and black‐blood aortic imaging, with improved image quality compared to T2Prep‐bSSFP and HASTE imaging, and comparable measurements for aortic wall and lumen dimensions. Evidence Level: 2. Technical Efficacy: Stage 2. [ABSTRACT FROM AUTHOR]

In this article, we propose a new U‐Net‐based approach for intraretinal cyst segmentation across different vendors that improve some of the challenges faced by previous deep‐based techniques. The proposed method has two main steps: (1) prior information embedding and input data adjustment, (2) the segmentation model. In the first step, we inject the information into the network in a way that overcomes some of the network limitations in receiving data and learning important contextual knowledge. And in the next step, we introduce a connection module between the encoder and decoder parts that transfers information more effectively from the encoder to the decoder. Two public datasets, namely, OPTIMA and KERMANY, are employed to evaluate the proposed method. The results show that the proposed method is an efficient vendor‐independent approach for the segmentation of intraretinal cystoid fluid with mean Dice values of 0.78 and 0.81 on the OPTIMA and KERMANY datasets, respectively. [ABSTRACT FROM AUTHOR]

Retinal vessel segmentation is an essential part of ocular disease diagnosis. However, due to complex vascular structure, large‐scale variations of retinal vessels, as well as inefficiency of vessel segmentation speed, accurate and fast automatic vessel segmentation for retinal images is still technically challenging. To tackle these issues, we present a lightweight multi‐scale‐aware network (LMSA‐Net) for retinal vessel segmentation. The network leverages the encoder‐decoder structure that was used in U‐Net. In the encoder, we propose a ghosted sandglass residual (GSR) block, aiming at greatly reducing the parameters and computational cost while obtaining richer semantic information. After that, a multi‐scale feature‐aware aggregation (MFA) module is designed to perceive multi‐scale semantic information for effective information extraction. Then, a global adaptive upsampling (GAU) module is proposed to guide the effective fusion of high‐ and low‐level semantic information in the decoder. Experiments are conducted on three public datasets, including DRIVE, CHASE_DB1, and STARE. The experimental results indicate the effectiveness of the LMSA‐Net, which can achieve better segmentation performance than other state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]

The current study aims to present differences between the pecten oculi of different avian species through morphologic, macroscopic, light, and electron microscopic examinations. The study is a comprehensive research on seven avian species (sparrowhawk, hawk, magpie, swan, heron, pheasant, duck). The right eyes of the animals utilized in the study were removed for light microscopic examination, whereas their left eyes were removed for electron microscopic and macroscopic examinations. Morphometric analyses, as well as stereo and light microscopic measurements, were carried out on the pecten oculi of the animals. Given all these data, it was determined that the height of the pecten oculi did not differ among the species in the study; however, the pecten oculi were larger in birds with the highest value compared to the other species in the macroscopic measurements. Also, the pecten oculi vessels were larger, and the number of melanocytes was higher in keen eyesight, raptor, and migratory birds with large bulbus oculi. All these data suggest that the pecten oculi not only supplies nutrient to the retina but also contributes to sharp vision during migration and hunting, UV absorption from sunlight, as well as preservation of intraocular equilibrium. [ABSTRACT FROM AUTHOR]

Purpose: To assess retinal findings in patients with severe carotid stenosis (CS) before and after carotid endarterectomy (CEA) compared to those in controls. Methods: This study is based on 70 patients (male 81%, mean age 69) scheduled for CEA in Helsinki University Hospital and 41 healthy nonmedicated controls (male 76%, mean age 68). Our examinations included fundus photographs. Semi‐automated software (Vesselmap, Imedos) served for evaluation of central retinal arterial equivalent (CRAE) and venular equivalent (CRVE), and arterio‐venous ratio (AVR) in both eyes. We assessed fundus photographs to subjectively grade microvascular abnormalities in the ipsilateral eyes including focal arteriolar narrowing and irregularities, arteriolar wall reflex, arterio‐venous crossing signs and arteriolar and venular tortuosity in the macula. Results: CRAE was similar in the ipsi‐and contralateral eyes of our patients, and similar to that of the controls both pre‐ and postoperatively. Preoperatively, we observed higher CRVE in the patients' ipsilateral than in their contralateral eyes (222 vs. 217 μm, p = 0.009), and likewise higher than in controls' eyes (222 vs. 214 μm, p = 0.024). CRVE decreased postoperatively in the patients' ipsilateral eyes (222 vs. 217 μm, p = 0.037). Among the microvascular abnormalities, arteriolar and venular tortuosity in the macula showed higher grades in the patients than in the controls preoperatively (p = 0.035 and p = 0.043), but not postoperatively (p = 0.15 and p = 0.10). Conclusions: CRVE decreased after CEA, showing that venules constrict after the mechanical hindrance of blood flow is removed. Higher grades in arteriolar and venular tortuosity in the macula, a potential ocular biomarker of CS, subsided after CEA. [ABSTRACT FROM AUTHOR]

Background: U‐Net and its variations have achieved remarkable performances in medical image segmentation. However, they have two limitations. First, the shallow layer feature of the encoder always contains background noise. Second, semantic gaps exist between the features of the encoder and the decoder. Skip‐connections directly connect the encoder to the decoder, which will lead to the fusion of semantically dissimilar feature maps. Purpose: To overcome these two limitations, this paper proposes a novel medical image segmentation algorithm, called feature‐guided attention network, which consists of U‐Net, the cross‐level attention filtering module (CAFM), and the attention‐guided upsampling module (AUM). Methods: In the proposed method, the AUM and the CAFM were introduced into the U‐Net, where the AUM learns to filter the background noise in the low‐level feature map of the encoder and the CAFM tries to eliminate the semantic gap between the encoder and the decoder. Specifically, the AUM adopts a top‐down pathway to use the high‐level feature map so as to filter the background noise in the low‐level feature map of the encoder. The AUM uses the encoder features to guide the upsampling of the corresponding decoder features, thus eliminating the semantic gap between them. Four medical image segmentation tasks, including coronary atherosclerotic plaque segmentation (Dataset A), retinal vessel segmentation (Dataset B), skin lesion segmentation (Dataset C), and multiclass retinal edema lesions segmentation (Dataset D), were used to validate the proposed method. Results: For Dataset A, the proposed method achieved higher Intersection over Union (IoU) (67.91±3.82%$67.91\pm 3.82\%$), dice (79.39±3.37%$79.39\pm 3.37\%$), accuracy (98.39±0.34%$98.39\pm 0.34\%$), and sensitivity (85.10±3.74%$85.10\pm 3.74\%$) than the previous best method: CA‐Net. For Dataset B, the proposed method achieved higher sensitivity (83.50%) and accuracy (97.55%) than the previous best method: SCS‐Net. For Dataset C, the proposed method had highest IoU (83.47±0.41%$83.47\pm 0.41\%$) and dice (90.81±0.34%$90.81\pm 0.34\%$) than those of all compared previous methods. For Dataset D, the proposed method had highest dice (average: 81.53%; retina edema area [REA]: 83.78%; pigment epithelial detachment [PED] 77.13%), sensitivity (REA: 89.01%; SRF: 85.50%), specificity (REA: 99.35%; PED: 100.00), and accuracy (98.73%) among all compared previous networks. In addition, the number of parameters of the proposed method was 2.43 M, which is less than CA‐Net (3.21 M) and CPF‐Net (3.07 M). Conclusions: The proposed method demonstrated state‐of‐the‐art performance, outperforming other top‐notch medical image segmentation algorithms. The CAFM filtered the background noise in the low‐level feature map of the encoder, while the AUM eliminated the semantic gap between the encoder and the decoder. Furthermore, the proposed method was of high computational efficiency. [ABSTRACT FROM AUTHOR]

Purpose: To report a case of feline bilateral corneal dermoids, associated with unilateral iris coloboma and bilateral choroido‐scleral colobomas in the same dorsolateral position, to describe retinographic and optical coherence tomography (OCT) characteristics, surgical outcome, and follow‐up. Animal studied: A 9‐month‐old domestic shorthaired cat in which a full ophthalmoscopic examination was performed for evaluation of dermoids resulting in a diagnosis of associated iris coloboma in one eye and posterior colobomas in both eyes. Procedures: Retinographies and OCT were performed under anesthesia to characterize the lesions of both fundi and allow surgical excision of the corneal dermoids. Results: Ophthalmoscopy and retinographies revealed oval lesions in the dorsolateral fundi of both eyes. The lesions precisely mirrored their respective dermoids' (10‐11 h OD and 1‐2 h OS) clock positions, lacked a tapetum lucidum and choroidal vessels, and featured thin retinal vessels plunging to a posterior plane of the fundus. OCT crossline scans demonstrated preservation of retinal thickness and morphological layering in the fundic colobomas leading to the conclusion that the colobomas were purely choroido‐scleral. The outcome of the surgical excision of the dermoids was satisfactory without hair recurrence and with acceptable corneal clarity making it possible to visualize the unilateral associated iris coloboma. Follow‐ups did not reveal any fundic evolution nor retinal detachment. Conclusions: Retinographies and OCT made possible the characterization of choroido‐scleral colobomas associated with corneal dermoids in this first reported case in a cat. We hypothesize that the recently described superior ocular sulcus might be the embryological link between these anomalies. [ABSTRACT FROM AUTHOR]

The primary objective of this research is to determine if the dynamic version of an averaged sub‐gradient (ASG) integral siding mode (ISM) controller can effectively handle the end‐effector acceleration tracking problem for a two‐link robotic arm avoiding singularity conditions. The proposed scheme solves a nonlinear extremum seeking problem that minimizes a non‐strictly convex function that depends on the acceleration tracking error defined by the difference between the desired acceleration trajectory and the cueing end‐effector acceleration of a two‐link robotic arm in a 2D‐plane. A second‐order differential equation drives the proposed dynamic ASG with nonlinear coefficients and a discontinuous right‐hand side term, which allows for the achievement of the necessary ISM regime in the presence of considerable errors in the mathematical model description. A special switch between modes allows the controller to track accelerations even in extreme positions (near to the active boundaries of the joint articulations) effectively. [ABSTRACT FROM AUTHOR]

Purpose: To identify the retinal vessel vasculature parameters associated with birdshot chorioretinopathy (BSCR). Methods: This retrospective observational study included 28 prevalent cases of BSCR with a median time from diagnosis of 6 years and 28 controls matched for age, arterial hypertension, diabetes and refraction. Forty‐five‐degree fundus images of both dilated eyes were acquired with a fundus camera (Canon CR‐2, Tokyo, Japan). The summary diameter of the arterial retinal vessels (central retinal artery equivalent, CRAE), venous retinal vessels (central retinal vein equivalent, CRVE), vascular tortuosity and fractal dimension (FD) were measured using VAMPIRE software. Retinal vasculitis was characterized using fluorescein angiography and active choroiditis using indocyanine green angiography. Results: At baseline, BSCR was associated with lower FD compared with matched controls (mean difference, −0.04; 95% confidence interval [CI], −0.06 to −0.02, p < 0.001). No other VAMPIRE parameters (CRAE, CRVE, arterial and venous tortuosity) differed. Among BSCR patients, retinal vein vasculitis was associated with higher CRAE (mean difference, 21 μ; 95% CI, 2.6–40, p = 0.03), venous tortuosity (geometric mean ratio, 1.79; 95% CI, 1.18–2.72, p = 0.007) and FD (mean difference, −0.04; 95% CI, −0.06 to −0.01, p = 0.007). Resolution of retinal vein vasculitis during follow‐up was paralleled by decrease in CRAE, CRVE and venous tortuosity values and increase in venous FD, respectively. Conclusion: BSCR is associated with lower FD value, suggesting that chronic retinal inflammation induces microvascular remodelling. Efficient treatment of retinal vasculitis may reverse changes in retinal vascular parameters. Changes in retinal vascular parameters could be potentially useful for assessing patients with BSCR disease. [ABSTRACT FROM AUTHOR]

Retinal vessel segmentation is an important computer vision task for eye retinopathy diagnosis. In the real scenarios, most datasets of source domain and target domain have distribution deviation, and the model often fails to generate accurate segmentation results due to the lack of data variation in single‐source domain, which damages the generalization ability to unseen target domains and may mislead doctors or artificial intelligence model in the following diseases diagnosis. Feature normalization is one feasible solution which can standardize data into uniform and stable distribution without additional data. However, the existing methods like batch normalization, uniform the data by global parameters. This leads to insufficient representation of important semantic information in the local region. To address this problem, the authors propose the spectral‐spatial normalization (SS‐Norm) module to enhance the generalization ability of the model. More specifically, the authors perform a discrete cosine transform (DCT) to decompose the feature into multiple frequency components and to analyze the semantic contribution degree of each component. By learning a spectral vector, the authors reweight the frequency components of features and therefore normalize the distribution in the spectral domain. Extensive experiments on six datasets prove the effectiveness of the authors' methods. [ABSTRACT FROM AUTHOR]

Climate change is transforming bioenergetic landscapes, challenging behavioral and physiological coping mechanisms. A critical question involves whether animals can adjust behavioral patterns and energy expenditure to stabilize fitness given reconfiguration of resource bases, or whether limits to plasticity ultimately compromise energy balance. In the Arctic, rapidly warming temperatures are transforming food webs, making Arctic organisms strong models for understanding biological implications of climate change‐related environmental variability. We examined plasticity in the daily energy expenditure (DEE) of an Arctic seabird, the little auk (Alle alle) in response to variability in climate change‐sensitive drivers of resource availability, sea surface temperature (SST) and sea ice coverage (SIC), and tested the hypothesis that energetic ceilings and exposure to mercury, an important neurotoxin and endocrine disrupter in marine ecosystems, may limit scope for plasticity. To estimate DEE, we used accelerometer data obtained across years from two colonies exposed to distinct environmental conditions (Ukaleqarteq [UK], East Greenland; Hornsund [HS], Svalbard). We proceeded to model future changes in SST to predict energetic impacts. At UK, high flight costs linked to low SIC and high SST drove DEE from below to above 4 × basal metabolic rate (BMR), a proposed energetic threshold for breeding birds. However, DEE remained below 7 × BMR, an alternative threshold, and did not plateau. Birds at HS experienced higher, relatively invariable SST, and operated above 4 × BMR. Mercury exposure was unrelated to DEE, and fitness remained stable. Thus, plasticity in DEE currently buffers fitness, providing resiliency against climate change. Nevertheless, modeling suggests that continued warming of SST may promote accelerating increases in DEE, which may become unsustainable. [ABSTRACT FROM AUTHOR]

Objective: To describe blindness, initially presenting as a unilateral condition and diagnosed as sudden acquired retinal degeneration syndrome (SARDS). Animals Studied: One Dachshund and four Maltese dogs presented with unilateral blindness, for which the results of general ophthalmic examinations were insufficient to explain the blindness. All dogs were spayed females. Results: Intraocular pressures were normal, and the optical media of the eyes were clear. Fundus appearances of the initially blind eyes were within normal variations, with slightly attenuated retinal blood vessels in some cases when compared with the sighted contralateral eyes. Electroretinography (ERG) amplitudes of the affected eyes were flat and reduced in the contralateral‐sighted eyes in four dogs. One dog underwent ERG after the blindness progressed bilaterally 8 days after initial presentation (despite topical steroid medication). Two dogs had no recheck visits, but phone call follow‐ups reported bilateral blindness 3 months later in one dog. One dog received no medication and retained vision in the contralateral eye until the last follow‐up (94 days later). One dog received systemic cyclosporine and steroid medications and maintained vision in the contralateral eye; however, regular ERG rechecks showed a trend of declining amplitude (448 days). In this dog, optical coherence tomography (OCT) showed different stages of disorganized retinal layers as well as different retinal thickness between the eyes. Conclusions: Despite normal‐looking fundi, ERG and OCT revealed different degrees of retinal changes between both eyes in this study. Eyes with vision might develop progressive blindness after a substantial amount of time in these presumed SARDS cases. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this pilot study was to assess oxygen saturation in retinal blood vessels in patients with monoclonal gammopathies (MGs). Methods: Thirty‐one patients with MGs (11 women and 20 men, mean age 65.9 ± 8.9 years) were enrolled during 2016–2020. The patients were diagnosed at the Haemato‐Oncology Department and subsequently examined at the Ophthalmology Department before initiating systemic therapy. All patients were subjected to automatic retinal oximetry (Oxymap ehf.) and had their fundus photographed (Topcon TRC‐50DX retinal camera). We assessed the association between retinal oxygen saturation (SatO2) – arterial SatO2, venous SatO2 and arterio‐venous (AV) difference—and MGs parameters: serum monoclonal immunoglobulin (M‐protein) level and serum immunoglobulin‐free light chains (FLC kappa and lambda), total protein, serum viscosity, haemoglobin, albumin, lactate dehydrogenase, C‐reactive protein, creatinine and serum calcium level. Hyperviscosity‐related retinopathy was also evaluated. Results: Statistical analysis showed a significant positive correlation (r = 0.462; p = 0.009) between the AV difference and the haemoglobin level. A significant, medium strong negative correlation was found between the AV difference and the serum levels of the monoclonal light lambda chains (r = −0.450; p = 0.011). Contrary to expectations, no statistically significant correlation was found between retinal oxygen saturation and the total protein or viscosity. Conclusion: This study found correlation between retinal oxygen saturation and certain parameters in the blood of patients with MGs. Increasing levels of monoclonal immunoglobulin seem to reduce oxygen absorption in retinal arterioles, resulting in a lower AV difference, particularly in patients with a high free light chain level. [ABSTRACT FROM AUTHOR]

Objective: To develop and validate a novel deep learning architecture to classify retinal vein occlusion (RVO) on color fundus photographs (CFPs) and reveal the image features contributing to the classification. Methods: The neural understanding network (NUN) is formed by two components: (1) convolutional neural network (CNN)‐based feature extraction and (2) graph neural networks (GNN)‐based feature understanding. The CNN‐based image features were transformed into a graph representation to encode and visualize long‐range feature interactions to identify the image regions that significantly contributed to the classification decision. A total of 7062 CFPs were classified into three categories: (1) no vein occlusion ("normal"), (2) central RVO, and (3) branch RVO. The area under the receiver operative characteristic (ROC) curve (AUC) was used as the metric to assess the performance of the trained classification models. Results: The AUC, accuracy, sensitivity, and specificity for NUN to classify CFPs as normal, central occlusion, or branch occlusion were 0.975 (± 0.003), 0.911 (± 0.007), 0.983 (± 0.010), and 0.803 (± 0.005), respectively, which outperformed available classical CNN models. Conclusion: The NUN architecture can provide a better classification performance and a straightforward visualization of the results compared to CNNs. [ABSTRACT FROM AUTHOR]

Background: Rapid and accurate segmentation of medical images can provide important guidance in the early stages of life‐threatening diseases. Purpose: However, fuzzy edges and high similarity with the background in images usually cause undersegmentation or oversegmentation. To solve these problems. Methods: We propose a novel edge features‐reinforcement (EFR) module that uses relative frequency changes before and after warping images to extract edge information. Then, the EFR module leverages deep features to guide shallow features to produce a band‐shaped edge attention map for reinforcing the edge region of all channels. We also propose a multiscale context exploration (MCE) module to fuse multiscale features and to extract channel and spatial correlations, which allows a model to focus on the parts that contribute most to the final segmentation. We construct EFR‐Net by embedding EFR and MCE modules on the encoder–decoder architecture. Results: We verify EFR‐Net's performance with four medical datasets: retinal vessel segmentation dataset DRIVE, endoscopic polyp segmentation dataset CVC‐ClinicDB, dermoscopic image dataset ISIC2018, and aortic true lumen dataset Aorta‐computed tomography (CT). The proposed model achieves Dice similarity coefficients (DSCs) of 81.61%, 92.87%, 89.87%, and 96.98% on DRIVE, CVC‐ClinicDB, ISIC2018, and Aorta‐CT, respectively, which are better than those of current mainstream methods. In particular, the DSC of polyp segmentation increased by 3.87%. Conclusion: Through quantitative and qualitative research, our method is determined to surpass current mainstream segmentation methods, and EFR modules can effectively improve the edge prediction effect of color images and CT images. The proposed modules are easily embedded in other encoder–decoder architectures, which has the potential to be applied and expanded. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the macular and circumpapillary retinal microvasculature across all stages of Leber hereditary optic neuropathy (LHON) using swept source optical coherence tomography angiography (OCTA). Methods: This prospective, multicentre, cross‐sectional, observational study analysed a total of 119 eyes of 60 patients with molecularly confirmed LHON across all stages and 120 eyes of 60 control subjects. Optical coherence tomography angiography maps centred on the fovea and optic disc were obtained to measure vessel densities (VDs) in the macular superficial (SCP) and deep (DCP) capillary plexuses, and the radial peripapillary capillary plexus (RPCP) respectively. Results: In asymptomatic eyes, only the SCP showed significant changes on average (B coefficient = −0.72, 95% CI = −1.34 to −0.10, p = 0.022) or in sectors representing the papillomacular bundle (PMB) (B coefficient = −1.17, 95% CI = −2.23 to −0.11, p = 0.031). However, in chronic eyes, the greatest magnitude of change was found in the temporal sector of the RPCP (B coefficient = −12.36, 95% CI = −14.49 to −10.23, p < 0.001). The RPCP showed the strongest correlations with visual acuity (VA, logarithm of the minimum angle of resolution; R = −0.677, p < 0.001) and structural parameters (R = 0.747, p < 0.001). Conclusions: Retinal VD changes in the circumpapillary region of the PMB appear later than in the macula but end up being more profound and correlate better with VA and structural parameters. Further studies are needed to assess the clinical utility of retinal VDs as potential LHON biomarkers. [ABSTRACT FROM AUTHOR]

In clinical medicine, the segmentation of blood vessels in retinal images is essential for subsequent analysis in clinical diagnosis. However, retinal images are often noisy and their vascular structure is relatively tiny, which poses significant challenges for vessel segmentation. To improve the performance of vessel segmentation, an improved model IterNet++ based on the architecture of IterNet is proposed. First, curvelet signal analysis is applied to enhance retinal images. Second, residual convolution (ResConv) blocks and guided filters are introduced to utilise the encoder features of previous iterations in the model to reduce overfitting. Third, offline hard‐sample mining is used to improve segmentation performance by utilising training samples with low segmentation accuracy as many possible on a few‐sample training set. In addition, a test‐time augmentation method is applied to testing samples in test dataset during inference. Extensive experiments show that this model achieves Dice scores of 0.8313, 0.8277, and 0.8372 on DRIVE, CHASE‐DB1, and STARE datasets, respectively, demonstrating the best performance compared with IterNet and other baseline models. [ABSTRACT FROM AUTHOR]

Purpose: The aim of the study was to obtain the values of oxygen saturation in retinal vessels and ophthalmic blood flow parameters in a healthy Caucasian population and assess whether the oximetry parameters are affected by the flow rate or the vascular resistance. Methods: The spectrophotometric retinal oximetry and colour Doppler imaging (CDI) of retinal vessels were successfully performed with 52 healthy subjects (average age 29.7 ± 5.6 years). The retinal oximeter simultaneously measures the wavelength difference of haemoglobin oxygen saturation in retinal arterioles and venules. The arteriolar and venular saturation in both eyes was measured. The peak systolic (PSV) end diastolic (EDV) velocities, resistive (RI) and pulsatility (PI) indices were obtained for both eyes using CDI in the ophthalmic artery. A paired t‐test and two sample t‐tests were used for statistical analyses. The correlation was assessed using the Pearson coefficient correlation. Results: The mean oxygen saturation level was 96.9 ± 3.0% for the retinal arterioles and 65.0 ± 5.1% for the retinal venules. The A‐V difference was 31.8 ± 4.6%. The mean of the measured haemodynamic parameters was PSV 46.6 ± 9.4 cm/s, EDV 12.0 ± 3.5 cm/s, PI 1.68 ± 0.38 and RI 0.74 ± 0.05. No significant difference in oxygen saturation and haemodynamic parameters was found between the left and the right eyes or the dominant and non‐dominant eye. The oximetry and ultrasound values were sex independent. The Pearson correlation coefficient demonstrated a significant yet weak negative correlation between A‐V difference and RI (r = −0.321, p = 0.020). Conclusions: A negative correlation between A‐V difference and resistance index was observed, suggesting that reduced oxygen consumption may reflect the increased vascular tone of the ophthalmic vessels, which is likely determined by autoregulatory mechanisms. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the association between cigarette smoking and retinal capillary plexus (RCP) using optical coherence tomography angiography (OCTA) and to examine whether potential vascular risk factors could impact their association. Methods: This is a cross‐sectional, community‐based study. The Jidong Eye Cohort Study included participants aged ≥18 years in the Jidong community (Tangshan city, northern China) from August 2019 to January 2020. All participants underwent comprehensive ophthalmic examination and completed detailed smoking questionnaires. Retinal vessel density in the superficial and deep RCP was automatically measured using OCTA. Results: Of the 2598 participants included in the study, 2026 (78.0%) never smoked and 572 (22.0%) had a history of smoking (494 [19.0%] current smokers and 78 [3.0%] former smokers). The median (interquartile range) age was 41 (34–52) years for the non‐smoking group and 45 (35–54.5) years for the smoking group. Multivariable analysis showed that smoking history is associated with a low deep RCP vessel density in the parafovea (β, −0.53; 95% confidence interval [CI], −0.82 to −0.24) and four quadrants. Increased smoking pack‐years were associated with reduced deep RCP vessel density in the parafovea (p for trend <0.001) and four quadrants. The significant interaction between diabetes and smoking only was found for superficial RCP vessel density in the parafovea (p for interaction = 0.014) and four quadrants except for the temporal quadrants. Conclusions: Cigarette smoking is an independent risk factor for reduced deep RCP vessel density. Our findings imply the potential detrimental effect of smoking on the occurrence of ocular diseases. [ABSTRACT FROM AUTHOR]

Purpose The aim of our study was to compare metabolic (oxygen saturation; %) and anatomical (diameter; μm) retinal vessel parameters of patients with central serous chorioretinopathy (CSC) to those of controls. Methods: In this prospective cross‐sectional cohort study, 72 eyes of patients with CSC were compared with 21 eyes of healthy controls. Of the 72 patients, 52 had chronic, nonactive CSC (subgroup nCSC) and 20 had active CSC (subgroup aCSC), according to activity on fluorescein angiography. Retinal vessel oximetry (RO) was performed using the Oxymap T1 oximeter. Oxygen saturation in all major peripapillary retinal arterioles (A‐SO2) and venules (V‐SO2) was measured, and their difference (A‐V SO2) was calculated. In addition, we evaluated the corresponding diameter in retinal arterioles (D‐A) and venules (D‐V). For statistical evaluation, ANOVA‐based linear mixed‐effects models were calculated (SPSS®; p < 0.05). Results: Central serous chorioretinopathy (CSC) patients had significantly higher A‐SO2 and V‐SO2 compared to that of controls (p = 0.031 and p = 0.018 respectively). Especially, the subgroup of aCSC patients showed significantly higher A‐SO2 and V‐SO2 values (p = 0.027 and p = 0.034, respectively). In addition, superotemporal and superonasal quadrant location showed significant interactions with A‐SO2 and V‐SO2 (p ≤ 0.03). Diameter in retinal arterioles (D‐A), an venules (D‐V) and A‐V SO2 findings showed no significant differences (p > 0.096). Conclusion: These data indicate that patients with CSC have altered metabolic function. The presence of disease activity showed the greatest influence on RO measurement, both compared to controls and to those with inactive chronic CSC disease. [ABSTRACT FROM AUTHOR]

Transfer RNA (tRNA)‐derived fragments are the non‐coding single‐stranded RNAs involved in several physiological and pathological processes. Herein, we investigated the role of tRF‐1020, a tRNA fragment, in diabetes‐induced retinal microvascular complications. The results showed that the levels of tRF‐1020 expression were down‐regulated in diabetic retinal vessels and retinal endothelial cells following high glucose or H2O2 stress. Overexpressing tRF‐1020 led to decreased endothelial cell viability, proliferation, migration, and tube formation and alleviated retinal vascular dysfunction as shown by decreased retinal acellular capillaries, vascular leakage, and inflammation. By contrast, tRF‐1020 silencing displayed the opposite effects. tRF‐1020 regulated endothelial angiogenic functions and retinal vascular dysfunction by targeting Wnt signalling. Moreover, the levels of tRF‐1020 expression were reduced in aqueous humour and vitreous samples of the patients with diabetic retinopathy. Collectively, tRF‐1020 is a potential target for the diagnosis and treatment of diabetic retinopathy. [ABSTRACT FROM AUTHOR]

Deep learning–based medical image segmentation is henceforth widely established as a powerful segmentation process. This article proposes a new U‐Net architecture based on a convolutional neural network for cytology image segmentation. This structure is more suitable to take into account pixel neighborhood in deconvolution. The goal is to develop an accurate segmentation method for white blood cells segmentation based on cells types features. This new proposed method yields a significant improvement compared to our previous work on the cytological medical dataset. In addition, the performance of the new architecture was also successfully tested on the Digital Retinal Image for Vessel Extraction databases benchmark. The images of this challenge are similar to our cytology image segmentation. Our approach achieved 25% relative improvement of the accuracy compared to the state‐of‐the‐art. [ABSTRACT FROM AUTHOR]