Vessel Edges Research Articles

Morphological data of the superior vena cava predicted by multiple linear regression equations

This study explores the main surface markers in the hemodialysis puncture catheterization pathway and the relative spatial position of venous vessels in the puncture results, providing data support for venous puncture. Seventy cadavers were studied; angles of vertical axis with superior vena cava, left and right brachiocephalic vein at the confluence and circumference of superior vena cava were the primary parameters. Bivariate correlation analysis was used to determine in relation to other body parameters, and a regression equation was also obtained. The angle of the vertical axis with the superior vena cava at the confluence was linearly correlated with the distances between the coracoid processes and the jugular notch to coracoid processes. The angle of the vertical axis with left brachiocephalic vein was linearly correlated with the thickness of thorax and the length from the upper pericardial margin to the confluence. The angle of the vertical axis with right brachiocephalic vein was correlated with clavicular midline length, acromial distance, intersection point of clavicle to clavicular midline, confluence to clavicular midline, superior edge of the blood vessels to superior edge of clavicle at the right intersection, and superior edge of pericardium to confluence. There is a linear relationship between the superior vena cava circumference and the distance from the superior vena cava to the aorta. A number of parameters influence angles of the vertical axis with superior vena cava, left and right brachiocephalic veins and the superior vena cava circumference which can provide guidance for intubation.

Platelet-Derived Growth Factor Subunit A Strengthens the Neurovascular Unit and Inhibits Retinal Vascular Regression Under Hyperoxic Conditions

Retinopathy of prematurity (ROP) is primarily caused by the exposure of preterm infants with underdeveloped blood vessels to high oxygen concentrations. This damages the astrocytes that promote normal vascular development, leading to avascularity, pathological neovascularization, and retinal detachment, and even blindness as the disease progresses. In this study, the aim was to investigate the differences in the characteristics of astrocytes and blood vessels between wild-type (WT) and genetically modified mice overexpressing platelet-derived growth factor subunit A (PDGF-A) in the retina immediately after high oxygen exposure, a protocol in the oxygen-induced retinopathy (OIR) model of ROP. Our results showed that PDGF-A mice exhibited an increased population of astrocytes and higher vascular density than WT mice and that PDGF-A strengthened the resistance to hyperoxic conditions. In the OIR model, PDGF-A mice had reduced avascular zone areas following hyperoxia exposure. Furthermore, immunostaining for NG2 and CD31 showed that pericytes tended to regress earlier than endothelial cells, particularly at the vessel edges in both WT and transgenic mice, indicating relatively higher susceptibility to hyperoxia-induced damage. These findings suggest that PDGF-A plays a crucial role in stabilizing retinal vessels and may serve as a novel therapeutic target for ROP, highlighting the potential significance of PDGF-A in the pathological mechanisms of retinal diseases.

Retinal artery/vein vessel segmentation and measurement for hypertension based on multi-level edge-guided network

Hypertension is a primary risk factor for the onset of cardiocerebrovascular diseases, leading to increased mortality. In the early stages of hypertension, changes in the diameter of the retinal arteries and veins are closely observed. Therefore, the automatic segmentation of these arteries and veins in retinal fundus images is crucial for hypertension monitoring. However, current deep learning based methods still fail to generate semantically consistent segmentation result, especially for narrow blood vessel segments. Moreover, the pixels near the vessel edge are prone to be missclassified by current methods, struggling to obtain edge-preserving segmentation results. These critical issues severely hinder downstream applications, which require accurate measurement of artery/vein diameter. First, to alleviate the issue of lacking semantic consistency for vessel segments, we propose a long-range spatial dependency modeling module to learn to model the long-range spatial dependency. On this basis, a multi-level edge guided spatial aggregation module is further presented to enhance the ability to accurately classify the edge pixels, generating edge-preserving results. Extensive experimental results on the widely used DRIVE dataset and our constructed dataset (HFC-50) show the superiority of the proposed model over state-of-the-art methods. Our method achieves balanced accuracy of 95.7% on DRIVE dataset, outperforming all state-of-art methods. Finally, in order to directly demonstrate the benefit of more accurate retinal artery/vein vessel segmentation for the measurement of the ratio of artery and vein, we quantitatively evaluate the computed artery/vein ratio for hypertension patients, with the artery/vein segmentation results generated by our proposed method.

Skeleton-guided multi-scale dual-coordinate attention aggregation network for retinal blood vessel segmentation

Deep learning plays a pivotal role in retinal blood vessel segmentation for medical diagnosis. Despite their significant efficacy, these techniques face two major challenges. Firstly, they often neglect the severe class imbalance in fundus images, where thin vessels in the foreground are proportionally minimal. Secondly, they are susceptible to poor image quality and blurred vessel edges, resulting in discontinuities or breaks in vascular structures. In response, this paper proposes the Skeleton-guided Multi-scale Dual-coordinate Attention Aggregation (SMDAA) network for retinal vessel segmentation. SMDAA comprises three innovative modules: Dual-coordinate Attention (DCA), Unbalanced Pixel Amplifier (UPA), and Vessel Skeleton Guidance (VSG). DCA, integrating Multi-scale Coordinate Feature Aggregation (MCFA) and Scale Coordinate Attention Decoding (SCAD), meticulously analyzes vessel structures across various scales, adept at capturing intricate details, thereby significantly enhancing segmentation accuracy. To address class imbalance, we introduce UPA, dynamically allocating more attention to misclassified pixels, ensuring precise extraction of thin and small blood vessels. Moreover, to preserve vessel structure continuity, we integrate vessel anatomy and develop the VSG module to connect fragmented vessel segments. Additionally, a Feature-level Contrast (FCL) loss is introduced to capture subtle nuances within the same category, enhancing the fidelity of retinal blood vessel segmentation. Extensive experiments on three public datasets (DRIVE, STARE, and CHASE_DB1) demonstrate superior performance in comparison to current methods. The code is available at https://github.com/wangwxr/SMDAA_NET.

Hyperreflective choroidal foci in diabetic eyes with and without macular edema: Novel insights on diabetic choroidopathy

Histopathologic studies of diabetic choroid suggest that diabetic choroidopathy is a key aspect secondary to diabetes. Recently, hyperreflective choroidal foci (HCF) have been introduced as novel optical coherence tomography (OCT) parameter. The aim of this study was to identify and quantify HCF in diabetic subjects with retinopathy, with or without diabetic macular edema (DME). Eighty-five diabetic subjects with different degrees of DR were enrolled: 37 without DME and 48 with DME. All subjects underwent full ophthalmologic examination including spectral domain optical coherence tomography (OCT). OCT images were analyzed to quantify and localize HCF. Each image was analyzed by two independent, masked examiners. OCT images showed that all subjects (100%) had HCF in the different layers of the choroid. The number of HCF was significantly higher in diabetics with DME versus those without DME (p < 0.0001). HCF showed variable size, shape and location inside the choroid. They were mainly located in choriocapillaris and Sattler's layer, on the edges of blood vessels. The intraobserver and interobserver agreement was almost perfect (ICC >0.9). This study suggests that hyperreflective foci in the choroid of subjects with DR may be accurately identified with structural OCT. Their number significantly increases with the progression of DME. These HCF may represent, as in the retina, a sign of infiltration of inflammatory cells (mainly migrating microglia) into the choroid, according to the hypothesis raised by Jerry Lutty. HCF may confirm in vivo the histopathologic findings suggesting that diabetic choroidopathy may be primarily a neuroinflammatory disorder.

Identifying diversity of patient anatomy through automated image analysis of clinical ultrasounds.

Central venous catheterization (CVC) carries inherent risks which can be mitigated through the use of appropriate ultrasound-guidance during needle insertion. This study aims to comprehensively understand patient anatomy as it is visualized during CVC by employing a semi-automated image analysis method to track the internal jugular vein and carotid artery throughout recorded ultrasound videos. The ultrasound visualization of 50 CVC procedures were recorded at Penn State Health Milton S. Hershey Medical Center. The developed algorithm was used to detect the vessel edges, calculating metrics such as area, position, and eccentricity. Results show typical anatomical variations of the vein and artery, with the artery being more circular and posterior to the vein in most cases. Notably, two cases revealed atypical artery positions, emphasizing the algorithm's precision in detecting anomalies. Additionally, dynamic vessel properties were analyzed, with the vein compressing on average to 13.4%of its original size and the artery expanding by 13.2%. This study provides valuable insights which can be used to increase the accuracy of training simulations, thus enhancing medical education and procedural expertise. Furthermore, the novel approach of employing automated data analysis techniques to clinical recordings showcases the potential for continual assessment of patient anatomy, which could be useful in future advancements.

HiDiffSeg: A hierarchical diffusion model for blood vessel segmentation in retinal fundus images

In medical diagnosis, achieving accurate segmentation of fundus vessels is crucial for understanding and identifying various anatomical structures. However, traditional algorithms often face challenges in achieving precise segmentation due to the poor quality of fundus images and the complex branching structure of vessels. To address this issue, we propose a novel approach called HiDiffSeg, which is a coarse-to-fine hierarchical diffusion model for blood vessel segmentation. Our method integrates an enhanced processing coarse-to-fine strategy for retinal fundus images into the denoising process of the diffusion model. The diffusion model is divided into two modules: the dual-guidance module (DGM) and the refinement-guidance module (RGM). The DGM takes the vascular skeleton image and the initial denoised vessel segmentation image generated by the vascular image enhancement model as conditions. Meanwhile, the RGM uses the fundus image as a condition to obtain more accurate results through iterative refinement. To emphasize the significance of vessel edges, we introduce a vessel enhancement module. By taking the original image as input, we generate a pixel-wise edge attention map, assigning greater importance to corresponding edge pixels. To the best of our knowledge, this is the first time a hierarchical diffusion model has been applied to fundus vessel segmentation. We demonstrate the accuracy of our method on three publicly available fundus retinal datasets (i.e., DRIVE, STARE, and CHASE_DB1) using evaluation metrics and compare it with eleven state-of-the-art fundus vessel segmentation methods.

Cross-patch feature interactive net with edge refinement for retinal vessel segmentation

Retinal vessel segmentation based on deep learning is an important auxiliary method for assisting clinical doctors in diagnosing retinal diseases. However, existing methods often produce mis-segmentation when dealing with low contrast images and thin blood vessels, which affects the continuity and integrity of the vessel skeleton. In addition, existing deep learning methods tend to lose a lot of detailed information during training, which affects the accuracy of segmentation. To address these issues, we propose a novel dual-decoder based Cross-patch Feature Interactive Net with Edge Refinement (CFI-Net) for end-to-end retinal vessel segmentation. In the encoder part, a joint refinement down-sampling method (JRDM) is proposed to compress feature information in the process of reducing image size, so as to reduce the loss of thin vessels and vessel edge information during the encoding process. In the decoder part, we adopt a dual-path model based on edge detection, and propose a Cross-patch Interactive Attention Mechanism (CIAM) in the main path to enhancing multi-scale spatial channel features and transferring cross-spatial information. Consequently, it improve the network’s ability to segment complete and continuous vessel skeletons, reducing vessel segmentation fractures. Finally, the Adaptive Spatial Context Guide Method (ASCGM) is proposed to fuse the prediction results of the two decoder paths, which enhances segmentation details while removing part of the background noise. We evaluated our model on two retinal image datasets and one coronary angiography dataset, achieving outstanding performance in segmentation comprehensive assessment metrics such as AUC and CAL. The experimental results showed that the proposed CFI-Net has superior segmentation performance compared with other existing methods, especially for thin vessels and vessel edges. The code is available at https://github.com/kita0420/CFI-Net.

LiViT-Net: A U-Net-like, lightweight Transformer network for retinal vessel segmentation

The intricate task of precisely segmenting retinal vessels from images, which is critical for diagnosing various eye diseases, presents significant challenges for models due to factors such as scale variation, complex anatomical patterns, low contrast, and limitations in training data. Building on these challenges, we offer novel contributions spanning model architecture, loss function design, robustness, and real-time efficacy. To comprehensively address these challenges, a new U-Net-like, lightweight Transformer network for retinal vessel segmentation is presented. By integrating MobileViT+ and a novel local representation in the encoder, our design emphasizes lightweight processing while capturing intricate image structures, enhancing vessel edge precision. A novel joint loss is designed, leveraging the characteristics of weighted cross-entropy and Dice loss to effectively guide the model through the task's challenges, such as foreground-background imbalance and intricate vascular structures. Exhaustive experiments were performed on three prominent retinal image databases. The results underscore the robustness and generalizability of the proposed LiViT-Net, which outperforms other methods in complex scenarios, especially in intricate environments with fine vessels or vessel edges. Importantly, optimized for efficiency, LiViT-Net excels on devices with constrained computational power, as evidenced by its fast performance. To demonstrate the model proposed in this study, a freely accessible and interactive website was established (https://hz-t3.matpool.com:28765?token=aQjYR4hqMI), revealing real-time performance with no login requirements.

Experience Using Gentian Violet-Free Dyes for Tissue Visualization.

Gentian violet ink is used as a skin marker in various surgical procedures, including neurosurgery. The dye is also used to visualize the edges of blood vessels during bypass surgery. However, gentian violet ink carries the risks of carcinogenicity and venous injury, which causes microvascular thrombosis. In this study, we compare the gentian violet-free dye C.I. Basic Violet 4 (BV4) and gentian violet. The usefulness, in terms of color, and formation of microvascular thrombosis in anastomosis were compared. We used the gentian violet-free dye in 20 cases involving 3 vascular anastomoses. The bone cutting lines on the bone surface, superior temporal artery, and middle cerebral artery were drawn using BV4 and gentian violet ink. The colors of BV4 and gentian violet ink were similar. No thrombus formation was observed at the vascular anastomosis when using BV4. BV4 can be used similarly to gentian violet ink. No adverse effects such as thrombus formation in microvascular anastomosis were experienced when BV4 was used.

CIDN: A context interactive deep network with edge-aware for X-ray angiography images segmentation

Accurate segmentation of X-ray angiography images is imperative for cardiovascular disease diagnosis. Despite significant strides in segmentation through deep learning, challenges persist in precisely delineating vessel edges. This paper introduces a Context Interactive Deep Network (CIDN) tailored for coronary vessel segmentation. CIDN integrates a Bio-inspired Attention Block (BAB) and a Multi-scale Interactive Block (MIB) to optimize the capture of spatial and edge information. The encoder-decoder facilitates optimal interaction between low-level features and high-level semantics. A compound loss function, amalgamating binary cross-entropy and active contour elasticity loss, enhances the recognition of intricate vessel edges. In experiments conducted on both public and private X-ray contrast images, the CIDN exhibits superior performance compared to the state-of-the-art Spatial Multi-scale Attention U-improved Network (SMAU-Net). Specifically, on dataset DCA1, the F1 value reached 0.7675, with an accuracy of 0.9795. Furthermore, on dataset JMA, the CIDN model achieved an F1 value of 0.8732, coupled with an accuracy of 0.9757. Accurate segmentation provides clinicians with precise vessel depictions, aiding in the identification of coronary stenosis and facilitating more informed diagnostic and therapeutic decisions.

PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation.

The structural morphology of mesenteric artery vessels is of significant importance for the diagnosis and treatment of colorectal cancer. However, developing automated vessel segmentation methods for this purpose remains challenging. Existing convolution-based segmentation methods have limitations in capturing long-range dependencies, while transformer-based models require large datasets, making them less suitable for tasks with limited training samples. Moreover, over-segmentation, mis-segmentation, and vessel discontinuity are common challenges in vessel segmentation tasks. To address these issues, we propose a parallel encoding architecture that combines transformers and convolutions to retain the advantages of both approaches. The model effectively learns position deviations and enhances robustness for small-scale datasets. Additionally, we introduce a vessel edge capture module to improve vessel continuity and topology. Extensive experimental results demonstrate the improved performance of our model, with Dice Similarity Coefficient and Average Hausdorff Distance scores of 81.64% and 7.7428, respectively.

Width measurement of retinal vessels using cubic spline fitting and extended edge‐searching mode

AbstractAbnormal 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.

TL-MSE2-Net: Transfer learning based nested model for cerebrovascular segmentation with aneurysms

Cerebrovascular (i.e., cerebral vessel) segmentation is essential for diagnosing and treating brain diseases. Convolutional neural network models, such as U-Net, are commonly used for this purpose. Unfortunately, such models may not be entirely satisfactory in dealing with cerebrovascular segmentation with tumors due to the following issues: (1) Relatively small number of clinical datasets from patients obtained through different modalities such as computed tomography (CT) and magnetic resonance imaging (MRI), leading to inadequate training and lack of transferability in the modeling; (2) Insufficient feature extraction caused by less attention to both convolution sizes and cerebral vessel edges. Inspired by the existence of similar features on cerebral vessels between normal subjects and patients, we propose a transfer learning strategy based on a pre-trained nested model called TL-MSE2-Net. This model uses one of the publicly available datasets for cerebrovascular segmentation with aneurysms. To address issue (1), our transfer learning strategy leverages a pre-trained model that uses a large number of datasets from normal subjects, providing a potential solution to the lack of sufficient clinical datasets. To tackle issue (2), we structure the pre-trained model based on 3D U-Net, comprising three blocks: ResMul, DeRes, and REAM. The ResMul and DeRes blocks enhance feature extraction by utilizing multiple convolution sizes to capture multiscale features, and the REAM block increases the weight of the voxels on the edges of the given 3D volume. We evaluated the proposed model on one small private clinical dataset and two publicly available datasets. The experimental results demonstrated that our MSE2-Net framework achieved an average Dice score of 70.81 % and 89.08 % on the two publicly available datasets, outperforming other state-of-the-art methods. Ablation studies were also conducted to validate the effectiveness of each block. The proposed TL-MSE2-Net yielded better results than MSE2-Net on a small private clinical dataset, with increases of 5.52 %, 3.37 %, 6.71 %, and 0.85 % for the Dice score, sensitivity, Jaccard index, and precision, respectively.

Color-Contrast Technique Using Fluorescein and Blue Marker to Maximize Visualization during Lymphaticovenous Anastomosis.

Lymphatic vessel wall and lumen visualization during anastomosis is challenging. Different techniques with variable efficacy have been described. Double-opposing color contrast is created using 10% fluorescein sodium, which stains lymphatic fluid yellow, causing a clear contrast to the blue marker-painted lymphatic wall, improving intralumen visualization during the anastomosis process. In this retrospective study, the authors evaluated the success rate of performing anastomosis between the side of the lymphatic vessel and the end of the vein (S-to-E LVA) in 281 patients. The LVA assessment showed mean lymphatic diameter of 0.44 ± 0.09 mm and mean vein diameter of 0.57 ± 0.14 mm with S-to-E success rate of 100% confirmed by postanastomosis indocyanine green lymphography. No adverse events were encountered. Fluorescein sodium was not used in 2 patients because of positive skin allergy test results. This method has the advantages of not needing an additional device, allowing clear visualization, and not staining the surrounding structures. This approach using opposing color contrast between fluorescent yellow and blue marker improved vessel edge identification, which translated into higher visualization and patency with 100% success rate in S-to-E LVA performance.

Digital subtraction angiography image segmentation based on multiscale Hessian matrix applied to medical diagnosis and clinical nursing of coronary stenting patients

ObjectiveTo introduce an improved method of digital subtraction angiography image segmentation based on a multi-scale Hessian matrix, to accurately segment vascular images before and after coronary stent implantation, which will have strong applications in medical diagnostics and clinical nursing of patients who underwent coronary stents implantation. MethodsFirstly, a vessel edge enhancement algorithm is proposed, which enhances the gradient of the vessel edge, which not only makes the obtained vessel edge smoother, but also enhances the visual effect of the intersection of vessels; secondly, introduces noise filtering based on morphology, which can detect and remove linear noise similar to blood vessels; finally, in view of the problem that each image in the DSA blood vessel sequence can only show part of the blood vessels, which is not conducive to observing the overall state of the blood vessels, a blood vessel sequence image fusion is designed to display the blood vessels in each image on the same image. The state of blood vessels can be understood as a whole in one image. ResultsCompared with the segmentation method in the literature, the blood vessel overlap rate increased by 0.0089, and the mis-segmentation rate decreased by 0.0334. In the quantitative analysis, the improved blood vessel segmentation algorithm in this paper improves the overlapping rate and reduces the mis-segmentation rate. ConclusionFor the comparison of visual effects, the blood vessels segmented by the algorithm in this paper have higher clarity, which is helpful for doctors and nurses to comprehensively synthesize blood vessel clinical information. There is a more objective image basis in the preoperative diagnosis and postoperative rehabilitation nursing of coronary angiographic stent implantation.

3D-shape formation of blood vessels based on computer aided design system

This paper proposes and tests a computerized approach for constructing a 3D model of blood vessels from angiogram images. The approach is divided into two steps, image features extraction and solid model formation. In the first step, image morphological operations and post-processing techniques are used for extracting geometrical entities from the angiogram image. These entities are the middle curve and outer edges of the blood vessel, which are then passed to a computer-aided graphical system for the second phase of processing. The system has embedded programming capabilities and pre-programmed libraries for automating a sequence of events that are exploited to create a solid model of the blood vessel. The gradient of the middle curve is adopted to steer the vessel’s direction, while the cross-sections of the blood vessel are formed as a sequence of circles lying in planes that are orthogonal to the gradients of the middle curves. The radii for the circles are estimated as a distance between the intersection points of the blood vessel edges with the orthogonal plane to the middle curve gradient. The system then uses these circles and the middle curve gradients to produce a solid volume that represents the 3D shape of the blood vessel. The method was tested and evaluated using different cases of angiogram images, and showed a reasonable agreement between the generated shapes and the tested images.

MS-CANet: Multi-Scale Subtraction Network with Coordinate Attention for Retinal Vessel Segmentation

Retinal vessel segmentation is crucial in the diagnosis of certain ophthalmic and cardiovascular diseases. Although U-shaped networks have been widely used for retinal vessel segmentation, most of the improved methods have insufficient feature extraction capability and fuse different network layers using element or dimension summation, leading to redundant information and inaccurate retinal vessel localization with blurred vessel edges. The asymmetry of small blood vessels in fundus images also increases the difficulty of segmenting blood vessels. To overcome these challenges, we propose a novel multi-scale subtraction network (MS-CANet) with residual coordinate attention to segment the vessels in retinal vessel images. Our approach incorporates a residual coordinate attention module during the encoding phase, which captures long-range spatial dependencies while preserving precise position information. To obtain rich multi-scale information, we also include multi-scale subtraction units at different perceptual field levels. Moreover, we introduce a parallel channel attention module that enhances the contrast between vessel and background, thereby improving the detection of marginal vessels during the decoding phase. We validate our proposed model on three benchmark datasets, namely DRIVE, CHASE, and STARE. The results demonstrate that our method outperforms most advanced methods under different evaluation metrics.

Transformer and convolutional based dual branch network for retinal vessel segmentation in OCTA images

Optical coherence tomography angiography (OCTA) enables detailed visualization of the vascular system. OCTA is of great significance for the diagnosis and treatment of many vision-related diseases. However, accurate retinal vessel segmentation is a great challenge due to obstacles such as low vessel edge visibility and high vessel complexity. We propose a novel OCTA retinal vessel segmentation method (ARP-Net) based on the Adaptive gated axial transformer (AGAT), Residual and Point repair modules. To reduce the impact of high vascular complexity on segmentation, we proposed a network composed of transformer and convolution branches to fuse the global and local information. Furthermore, considering the high computation of transformer, we propose an AGAT in the transformer branch. Finally, the low visibility of regions such as vessel edge in OCTA images makes the prediction of the network in these regions difficult. Therefore, we also propose a point repair module to re-predict these regions. We have performed experiments on two public OCTA vessel segmentation datasets and achieved better results than the latest state-of-the-art methods.

SPNet: A novel deep neural network for retinal vessel segmentation based on shared decoder and pyramid-like loss

Segmentation of retinal vessel images is critical to the diagnosis of retinopathy. Recently, convolutional neural networks have shown significant ability to extract the blood vessel structure. However, it remains challenging to refined segmentation for the capillaries and the edges of retinal vessels due to thickness inconsistencies and blurry boundaries. In this paper, we propose a novel deep neural network for retinal vessel segmentation based on shared decoder and pyramid-like loss (SPNet) to address the above problems. Specifically, we introduce a decoder-sharing mechanism to capture multi-scale semantic information, where feature maps at diverse scales are decoded through a sequence of weight-sharing decoder modules. Also, to strengthen characterization on the capillaries and the edges of blood vessels, we define a residual pyramid architecture which decomposes the spatial information in the decoding phase. A pyramid-like loss function is designed to compensate possible segmentation errors progressively. Experimental results on public benchmarks show that the proposed method outperforms the backbone network and most state-of-the-art methods, especially in the regions of the capillaries and the vessel contours. In addition, performances on cross-datasets verify that SPNet shows stronger generalization ability.