Segmental Branches Research Articles

Edge-aware object pixel-level representation tracking

Recently, there has been a trend in tracking to use more refined segmentation mask instead of coarse bounding box to represent the target object. Some trackers proposed segmentation branches based on the tracking framework and maintain real-time speed. However, those trackers use a simple FCNs structure and lack of the edge information modeling. This makes performance quite unsatisfactory. In this paper, we propose an edge-aware segmentation network, which uses the complementarity between target information and edge information to provide a more refined representation of the target. Firstly, We use the high-level features of the tracking backbone network and the correlation features of the classification branch of the tracking framework to fuse, and use the target edge and target segmentation mask for simultaneous supervision to obtain an optimized high-level feature with rough edge information and target information. Secondly, we use the optimized high-level features to guide the low-level features of the tracking backbone network to generate more refined edge features. Finally, we use the refined edge features to fuse with the target features of each layer to generate the final mask. Our approach has achieved leading performance on recent pixel-wise object tracking benchmark VOT2020 and segmentation datasets DAVIS2016 and DAVIS2017 while running on 47 fps. Code is available at https://github.com/TJUMMG/EATtracker.

Wellens Syndrome on the basis of critical occlusion of the medial segment of the anterior ventricultural branch of the left coronary artery (clinical type)

The aim of the work is to determine the peculiarities of the clinical course of Wellens syndrome due to the presence of critical occlusion of the medial segment of the anterior interventricular coronary artery bifurcation.Material and Methods. Clinical manifestations of the disease, including dynamics of ECG changes; coronary angiography data; results of echocardiographic and laboratory tests.Results. This article presents the peculiarities of the clinical course of Wellens syndrome in patient L., 70, who met the currently recognized diagnostic criteria of the syndrome, but the critical occlusion of the anterior interventricular branch of the left coronary artery was localized in its medial segment. Along with this, there was hemodynamically irrelevant stenosis of the proximal coronary artery section (60%). Electrocardiographic changes were consistent with pattern B (deeply inverted T waves in V2-V6). Сonditional normalization of the T waves, indicating the absence of repeated coronary artery occlusion after spontaneous reperfusion. Stenting of the anterior interventricular coronary artery vein of the left coronary artery prevented the possible occurrence of “stuttering” pattern and the occurrence of myocardial infarction.Conclusions. Acute coronary syndrome with complete coronary artery occlusion is often accompanied by spontaneous reperfusion, which contributes to the relief of chest pain and the appearance of characteristic T waves on ECG and is the basis of the so-called Wellens syndrome. The most frequent localization of coronary artery occlusion is the proximal segment of the anterior interventricular branch of the left coronary artery, but we witnessed the occlusion in the medial segment of these coronary arteries. Immediate coronary angiography with stenting of the occluded artery will prevent the occurrence of myocardial infarction in the patient.

Anatomy of the coronary arteries and localization of coronary atherosclerosis in siblings with coronary heart disease

There is no doubt that there is a similar genetic predisposition in close relatives to the development of cardiovascular disease, while the features of coronary anatomy and possible commonality of pathological process in close relatives remain beyond research interest.
 Aim. To study the anatomy of the coronary arteries, the nature and severity of coronary lesion in pairs of male siblings suffering from coronary artery disease, in comparison with control pairs of patients who are not closely related.
 Materials and methods. This retrospective study included 87 male patients. The main group (58 patients or 29 couples) consisted of siblings with a horizontal hereditary burden for coronary artery disease. The control subgroup included 29 big brothers of the main group and another 29 patients who unrelated to them. To assess the degree of anatomy identity of the coronary bed in pairs of patients, we used a comparative characteristic of 25 segments of the coronary arteries (SyntaxScore). If there was similarity in anatomy and in the severity of the atherosclerotic process, each segment was assigned one point. In terms of the ratio of the total scores to the total number of segments (n=25), each pair of patients was assigned anatomical and atherosclerotic identity indexes.
 Results. According to the results of the study, the average indices of anatomical and atherosclerotic identity indexes were significantly higher in pairs of sibs compared to control pairs of patients (0.92 versus 0.88, p=0.008 and 0.92 versus 0.76, p0.001 respectively). When analyzing atherosclerotic lesions of the coronary arteries in pairs of sibs, a statistically significant coincidence of lesions in a number of segments of the coronary arteries was revealed, with the highest agreement in the left main coronary artery (=0.869, p0.001) and in the proximal segments of the main branches of the coronary arteries: left anterior descending artery (=0.786, p0.001) and right coronary artery (=0.812, p0.001). In the group of control pairs, such regularities were not revealed.
 Conclusion. In siblings with horizontal hereditary burden for coronary artery disease along the sibling line, the average anatomical and atherosclerotic identity indexes are significantly higher compared to control pairs. In pairs of siblings, there is the highest and most reliable probability of developing coronary atherosclerosis in the left main coronary artery and in the proximal segments of the left anterior descending artery and right coronary artery.

929: ENDOBRONCHIAL METASTASIS FROM OSTEOSARCOMA: TREATMENT WITH CRYOTHERAPY AND ARGON PLASMA COAGULATION

Introduction: Osteosarcoma is the most common primary malignant tumor of the bone. Pulmonary metastasis is frequently seen as a relapse in approximately 80% of patients; however, endobronchial metastasis is very rare. We present a case of endobronchial osteosarcoma in a young man. Description: A 20-year-old male non-smoker with a history of treatment-refractory metastatic osteosarcoma presented with shortness of breath, dry cough, and episodes of hemoptysis for a few days. On presentation, vital signs were significant for tachycardia (HR 113/min) and tachypnea (RR 24/min). Chest CT scan showed no pulmonary embolism but revealed bilateral pulmonary and pleural metastases with severe narrowing of the right upper lobe pulmonary artery segmental branches secondary to metastases. At fiberoptic bronchoscopy, there was complete obstruction of the right upper lobe apical and posterior segments by a dark red vascular appearing tumor. Cryotherapy was performed to destroy abnormal tissue, along with balloon occlusion. Coagulation using an Argon Plasma Coagulation (APC) was performed in the lung. Post-treatment findings showed improved lumen size. The patient tolerated the procedure well and had remarkable improvement in symptoms. Biopsy of the endobronchial tumor showed metastatic sarcoma, similar to the patient’s prior high-grade pleomorphic sarcoma. Discussion: Airway obstruction may be extrinsic, intrinsic, or mixed; fixed or dynamic; and benign or malignant. It may present with a variety of symptoms ranging from mild shortness of breath to respiratory failure. There are few published case reports of endobronchial obstruction secondary to osteosarcoma. Bronchoscopy (either rigid or flexible) is always necessary for assessing airway obstruction. Dilation is immediately effective for intrinsic and extrinsic compression, but results are typically not sustained. Cryotherapy relies on repeated freeze-thaw cycles using extreme cold (below –40°C) to destroy tissue and tumor and can be used via both rigid and flexible bronchoscopes. APC is a mode of noncontact electrocoagulation that is increasingly used as an alternative to contact electrocautery and noncontact laser therapy. Often, the best therapeutic approach appears to include a combination of several treatment approaches as demonstrated in our patient.

Fully automatic tumor segmentation of breast ultrasound images with deep learning.

Breast ultrasound (BUS) imaging is one of the most prevalent approaches for the detection of breast cancers. Tumor segmentation of BUS images can facilitate doctors in localizing tumors and is a necessary step for computer-aided diagnosis systems. While the majority of clinical BUS scans are normal ones without tumors, segmentation approaches such as U-Net often predict mass regions for these images. Such false-positive problem becomes serious if a fully automatic artificial intelligence system is used for routine screening. In this study, we proposed a novel model which is more suitable for routine BUS screening. The model contains a classification branch that determines whether the image is normal or with tumors, and a segmentation branch that outlines tumors. Two branches share the same encoder network. We also built a new dataset that contains 1600 BUS images from 625 patients for training and a testing dataset with 130 images from 120 patients for testing. The dataset is the largest one with pixel-wise masks manually segmented by experienced radiologists. Our code is available at https://github.com/szhangNJU/BUS_segmentation. The area under the receiver operating characteristic curve (AUC) for classifying images into normal/abnormal categories was 0.991. The dice similarity coefficient (DSC) for segmentation of mass regions was 0.898, better than the state-of-the-art models. Testing on an external dataset gave a similar performance, demonstrating a good transferability of our model. Moreover, we simulated the use of the model in actual clinic practice by processing videos recorded during BUS scans; the model gave very low false-positive predictions on normal images without sacrificing sensitivities for images with tumors. Our model achieved better segmentation performance than the state-of-the-art models and showed a good transferability on an external test set. The proposed deep learning architecture holds potential for use in fully automatic BUS health screening.

A framework for computing angle of progression from transperineal ultrasound images for evaluating fetal head descent using a novel double branch network.

Background: Accurate assessment of fetal descent by monitoring the fetal head (FH) station remains a clinical challenge in guiding obstetric management. Angle of progression (AoP) has been suggested to be a reliable and reproducible parameter for the assessment of FH descent. Methods: A novel framework, including image segmentation, target fitting and AoP calculation, is proposed for evaluating fetal descent. For image segmentation, this study presents a novel double branch segmentation network (DBSN), which consists of two parts: an encoding part receives image input, and a decoding part composed of deformable convolutional blocks and ordinary convolutional blocks. The decoding part includes the lower and upper branches, and the feature map of the lower branch is used as the input of the upper branch to assist the upper branch in decoding after being constrained by the attention gate (AG). Given an original transperineal ultrasound (TPU) image, areas of the pubic symphysis (PS) and FH are firstly segmented using the proposed DBSN, the ellipse contours of segmented regions are secondly fitted with the least square method, and three endpoints are finally determined for calculating AoP. Results: Our private dataset with 313 transperineal ultrasound (TPU) images was used for model evaluation with 5-fold cross-validation. The proposed method achieves the highest Dice coefficient (93.4%), the smallest Average Surface Distance (6.268 pixels) and the lowest AoP difference (5.993°) by comparing four state-of-the-art methods. Similar results (Dice coefficient: 91.7%, Average Surface Distance: 7.729 pixels: AoP difference: 5.110°) were obtained on a public dataset with >3,700 TPU images for evaluating its generalization performance. Conclusion: The proposed framework may be used for the automatic measurement of AoP with high accuracy and generalization performance. However, its clinical availability needs to be further evaluated.

Anatomical Evidence for Parasympathetic Innervation of the Renal Vasculature and Pelvis.

The kidneys critically contribute to body homeostasis under the control of the autonomic nerves, which enter the kidney along the renal vasculature. Although the renal sympathetic and sensory nerves have long been confirmed, no significant anatomic evidence exists for renal parasympathetic innervation. We identified cholinergic nerve varicosities associated with the renal vasculature and pelvis using various anatomic research methods, including a genetically modified mouse model and immunostaining. Single-cell RNA sequencing (scRNA-Seq) was used to analyze the expression of AChRs in the renal artery and its segmental branches. To assess the origins of parasympathetic projecting nerves of the kidney, we performed retrograde tracing using recombinant adeno-associated virus (AAV) and pseudorabies virus (PRV), followed by imaging of whole brains, spinal cords, and ganglia. We found that cholinergic axons supply the main renal artery, segmental renal artery, and renal pelvis. On the renal artery, the newly discovered cholinergic nerve fibers are separated not only from the sympathetic nerves but also from the sensory nerves. We also found cholinergic ganglion cells within the renal nerve plexus. Moreover, the scRNA-Seq analysis suggested that acetylcholine receptors (AChRs) are expressed in the renal artery and its segmental branches. In addition, retrograde tracing suggested vagus afferents conduct the renal sensory pathway to the nucleus of the solitary tract (NTS), and vagus efferents project to the kidney. Cholinergic nerves supply renal vasculature and renal pelvis, and a vagal brain-kidney axis is involved in renal innervation.

BANet: Boundary-Assistant Encoder-Decoder Network for Semantic Segmentation

Recently, boundary information has gained great attraction for semantic segmentation. This paper presents a novel encoder-decoder network, called BANet, for accurate semantic segmentation, where boundary information is employed as an additional assistance for producing more consistent segmentation outputs. BANet is composed of three components: the pre-trained backbone using dilated-ResNet101, semantic flow branch (SFB) and boundary flow branch (BFB) for semantic segmentation and boundary detection, respectively. More specifically, to delineate more accurate object shapes and boundaries, a global attention block (GAB) is designed in SFB as global guidance for high-level feature. On the other hand, BFB directly extracts features on boundaries, avoiding the unexpected interference from the non-boundary parts. Finally, we adopt a joint loss function to further optimize the segmentation results and boundary outputs synchronously. Moreover, compared with previous state-of-the-art methods, e.g., non-local block and ASPP module, our BFB leverages detection accuracy and computational efficiency in a lightweight fashion. To evaluate BANet, we have conducted extensive experiments on several semantic segmentation datasets: Cityscapes, PASCAL Context, and ADE20K. The experimental results show that, with the aid of boundary information, BANet is able to produce more consistent segmentation predictions with accurately delineated object shapes and boundaries, leading to the state-of-the-art performance on Cityscapes, and competitive results on PASCAL Context and ADE20K with respect to recent semantic segmentation networks.

Local application of heparin ointment as a trigger factor for the heparin-induced thrombocytopenia

Heparin-induced thrombocytopenia (GIT) is a paradoxical reaction of platelets to heparin (their aggregation and destruction), potentially dangerous by the development of venous and arterial thrombosis, less often bleeding. There are serious problems of timely diagnosis of this complication. They are due to both insufficient awareness of specialists and the presence of a negative psychological attitude. The article presents a clinical case of treatment of a 57-year-old patient who was prescribed local application of heparin ointment for catheter thrombosis of the superficial veins of the left upper limb. After 8 days, the patient developed a clinical and instrumental picture of occlusive thrombosis of the right brachial vein. The patient was prescribed enoxaparin 0.6 x 2 times a day subcutaneously. After 4 days since the appointment of enoxaparin, a sharp deterioration in the patient’s condition: lower paraplegia with impaired pelvic organ function developed. There was a decrease in platelet levels by more than 2 times to 49x109/l. The probability of developing GIT on the «4T» scale is 8 points (high probability of GIT). Despite the treatment, the patient’s condition progressively worsened with the development of thrombosis of the infrarenal aorta, both common iliac arteries, massive trunk thromboembolism, segmental and subsegmental branches of the left pulmonary artery, intracerebral hemorrhage with a breakthrough into the ventricular system. Conclusion: heparin-induced thrombocytopenia, a rare but extremely dangerous complication observed in patients receiving heparin. The key to preventing the catastrophic consequences of GIT is awareness, alertness of clinicians and the use of modern algorithms for the diagnosis and treatment of GIT.

Robust Medical Image Colorization with Spatial Mask-Guided Generative Adversarial Network.

Color medical images provide better visualization and diagnostic information for doctors during clinical procedures than grayscale medical images. Although generative adversarial network-based image colorization approaches have shown promising results, in these methods, adversarial training is applied to the whole image without considering the appearance conflicts between the foreground objects and the background contents, resulting in generating various artifacts. To remedy this issue, we propose a fully automatic spatial mask-guided colorization with generative adversarial network (SMCGAN) framework for medical image colorization. It generates colorized images with fewer artifacts by introducing spatial masks, which encourage the network to focus on the colorization of the foreground regions instead of the whole image. Specifically, we propose a novel spatial mask-guided method by introducing an auxiliary foreground segmentation branch combined with the main colorization branch to obtain the spatial masks. The spatial masks are then used to generate masked colorized images where most background contents are filtered out. Moreover, two discriminators are utilized for the generated colorized images and masked generated colorized images, respectively, to assist the model in focusing on the colorization of foreground regions. We validate our proposed framework on two publicly available datasets, including the Visible Human Project (VHP) dataset and the prostate dataset from NCI-ISBI 2013 challenge. The experimental results demonstrate that SMCGAN outperforms the state-of-the-art GAN-based image colorization approaches with an average improvement of 8.48% in the PSNR metric. The proposed SMCGAN can also generate colorized medical images with fewer artifacts.

SEG-ESRGAN: A Multi-Task Network for Super-Resolution and Semantic Segmentation of Remote Sensing Images

The production of highly accurate land cover maps is one of the primary challenges in remote sensing, which depends on the spatial resolution of the input images. Sometimes, high-resolution imagery is not available or is too expensive to cover large areas or to perform multitemporal analysis. In this context, we propose a multi-task network to take advantage of the freely available Sentinel-2 imagery to produce a super-resolution image, with a scaling factor of 5, and the corresponding high-resolution land cover map. Our proposal, named SEG-ESRGAN, consists of two branches: the super-resolution branch, that produces Sentinel-2 multispectral images at 2 m resolution, and an encoder–decoder architecture for the semantic segmentation branch, that generates the enhanced land cover map. From the super-resolution branch, several skip connections are retrieved and concatenated with features from the different stages of the encoder part of the segmentation branch, promoting the flow of meaningful information to boost the accuracy in the segmentation task. Our model is trained with a multi-loss approach using a novel dataset to train and test the super-resolution stage, which is developed from Sentinel-2 and WorldView-2 image pairs. In addition, we generated a dataset with ground-truth labels for the segmentation task. To assess the super-resolution improvement, the PSNR, SSIM, ERGAS, and SAM metrics were considered, while to measure the classification performance, we used the IoU, confusion matrix and the F1-score. Experimental results demonstrate that the SEG-ESRGAN model outperforms different full segmentation and dual network models (U-Net, DeepLabV3+, HRNet and Dual_DeepLab), allowing the generation of high-resolution land cover maps in challenging scenarios using Sentinel-2 10 m bands.

The effect of subbranch for the quantification of local hemodynamic environment in the coronary artery: a computed tomography angiography–based computational fluid dynamic analysis

Abstract Background Hemodynamic parameters derived from computed tomography angiography–based computational fluid dynamics (CFD) analysis have been widely used for clinical decision-making and researches to assess the vulnerability of atherosclerotic plaques and explain the initialization and development of atherosclerosis. Subbranches in the CFD model might affect the accuracy of hemodynamic parameters, but the effectiveness has been least quantified. Methods A coronary artery baseline model was generated with focal stenosis at the proximal left anterior descending artery. Nineteen comparing models were created by systematically removing various subbranches to examine the changes in hemodynamic parameters, including time-averaged pressure (TAP), time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), and particle relative residence time (RRT). Changes in these parameters were assessed quantitatively around the stenosis and near the region where subbranches were removed. Results The removal of subbranches caused a significant change in outflow rate, and there was generally a decrease in all CFD parameters in the regions of interest with a decrease in outflow rate. The subbranch removal had a significant impact on the calculation of TAWSS, OSI, and RRT, whereas TAP was insensitive to the removal with approximately 0.25% variation in all 19 models. The local effect from removing branch segments generally became negligible after 5 diameters away from the cutting-off position, but the decrease could be affected by other factors, such as a large curvature. Conclusion The outflow rate is a dominant factor for the calculation of TAP, TAWSS, OSI, and RRT. Removal of subbranches has a minor effect on the TAP calculation, but its effect is considerable on the TAWSS, OSI, and RRT. The effect of subbranch removal is limited in a region with 5 local diameters.

Weakly Supervised Instance Segmentation via Category-aware Centerness Learning with Localization Supervision

Deep convolutional neural networks (DCNN) trained with pixel-level segmentation masks achieve high performance in the task of instance segmentation. The difficulty of acquiring such annotation limits the application and popularization of the DCNN-based approaches. To address the issue, a weakly supervised approach is proposed in the paper which performs instance segmentation only with the supervision of bounding box or coarse localization annotation. A novel DCNN model is constructed which consists of two branches: the centerness branch and the segmentation branch. The former branch is to learn the semantically spatial importance over the areas of object instances under the localization supervision. Object proposals with exact boundaries are automatically generated and are then ranked under the guidance of the output of the centerness branch. The most matched instance proposal is assigned to each object, which is then used to supervise the segmentation branch. The losses are calculated by both the outputs of the two branches and the entire DCNN model is trained end-to-end. Experiments are extensively conducted to verify the effectiveness. With the supervision of precise bounding box annotation, our approach achieves state-of-the-art (SOTA) accuracy in the comparison with recent related works. And in the case of coarse localization annotation, our approach only deduces a slight reduction in accuracy, which significantly outperforms other approaches. The excellent performance demonstrates that our approach would be helpful to further alleviate the workload of image annotation while maintaining competitive accuracy.

Main stem intervention-Is one stent better than two after all?

The current European Society of Cardiology (ESC) guideline recommendations give equal consideration to percutaneous coronary interventions (PCI) and bypass surgery for the treatment of main stem stenosis. Because of the anatomical setting, interventional treatment of main stem stenosis represents acomplex PCI procedure. For this reason, extensive planning of the PCI with the use of intravascular imaging is necessary to assess the extent of calcification in the main stem itself and also in the proximal vascular segment of the anterior interventricular branch (left anterior descending, LAD) and the circumflex branch (RCX). Furthermore, an assessment of the vessel diameter is necessary for planning the PCI as well as for stent selection. Especially in the case of severely calcified lesions, careful and routine dissection of the lesion to be treated is essential. For this purpose, so-called cutting balloons, scoring balloons or debulking devices (rotational atherectomy, orbital atherectomy) can be used. In the case of amain stem stenosis involving the bifurcation, the choice of stent strategy depends on the extent of calcification of the daughter vessels (LAD and RCX), as well as the complexity of the bifurcation stenosis. The superiority of the so-called double kissing (DK) crush technique over asingle stent strategy could be confirmed in arandomized trial and should be used in main stem bifurcation stenoses with pronounced calcification of the proximal segments of the LAD and RCX (stenosis length > 10 mm, and > 70% diameter stenosis). The already established 2‑stent strategies for the treatment of bifurcation stenosis could be modified or further developed. In preliminary studies the advantage of the so-called DK nano-crush technique and the DK culotte technique could be shown, whereby these initial results still have to be confirmed by larger randomized studies.

Reconstruction of tree branching structures from UAV-LiDAR data

The reconstruction of tree branching structures is a longstanding problem in Computer Graphics which has been studied over several data sources, from photogrammetry point clouds to Terrestrial and Aerial Laser Imaging Detection and Ranging technology. However, most data sources present acquisition errors that make the reconstruction more challenging. Among them, the main challenge is the partial or complete occlusion of branch segments, thus leading to disconnected components whether the reconstruction is resolved using graph-based approaches. In this work, we propose a hybrid method based on radius-based search and Minimum Spanning Tree for the tree branching reconstruction by handling occlusion and disconnected branches. Furthermore, we simplify previous work evaluating the similarity between ground-truth and reconstructed skeletons. Using this approach, our method is proved to be more effective than the baseline methods, regarding reconstruction results and response time. Our method yields better results on the complete explored radii interval, though the improvement is especially significant on the Ground Sampling Distance In terms of latency, an outstanding performance is achieved in comparison with the baseline method.

Interventional Imaging Roadmap to Successful Balloon Pulmonary Angioplasty for Chronic Thromboembolic Pulmonary Hypertension.

Balloon pulmonary angioplasty (BPA) is an evolving treatment modality for patients with chronic thromboembolic pulmonary hypertension (CTEPH) who are not candidates for pulmonary endarterectomy. Although several imaging modalities currently exist for evaluating CTEPH, their individual use, specifically in the clinical practice of BPA, has not been well described. In this article, we provide a preprocedural, intraprocedural, and postprocedural interventional imaging roadmap for safe and effective BPA performance in routine clinical practice. Preprocedural assessment includes transthoracic echocardiography for right ventricular assessment, ventilation/perfusion scan to identify pulmonary segments with the highest degree of hypoperfusion, cross-sectional chest imaging excluding alternative causes of mismatched defects and providing anatomic and perfusion imaging concurrently, and nonselective invasive pulmonary angiography for risk stratification of individual lesion subtypes. Intraprocedural assessment includes subselective segmental angiography (SSA) for delineating segmental and subsegmental branch anatomy, lesion identification, and vessel sizing. Intravascular ultrasound and optical coherence tomography serve as adjunctive intraprocedural tools for more accurate vessel sizing and lesion characterization when SSA alone is insufficient. Postprocedural considerations include chest radiography to monitor for immediate postprocedure complications and echocardiography for the interval assessment of the right ventricle on longer-term follow-up.

Coil Embolization of Variant Hepatic Arteries During Percutaneous Arterial Port Catheter Placement for Intraarterial Chemotherapy: Analysis of Intrahepatic Perfusion Redistribution and Treatment Efficacy.

The purpose of this study was to analyze the intrahepatic perfusion redistribution after embolization of hepatic arterial variants during percutaneous arterial port catheter placement as well as to investigate the treatment efficacy of intraarterial chemotherapy in perfusion redistribution-dependent compared to redistribution-independent liver areas. This retrospective study included 62 patients (67.7% males, mean age of 56 ± 12years). A replaced left hepatic artery was encountered in 36/62 (58.1%), a replaced right hepatic artery in 19/62 (30.6%) and a replaced left and right hepatic artery in 7/62 of patients (11.3%), respectively. Subjective perfusion analysis was performed on digital subtracted angiography and computed tomography (CT)/cone-beam computed tomography (CBCT) images evaluating the visibility of the main, segmental and subsegmental branches of the embolized variant hepatic artery, re-perfused from intrahepatic arterial anastomoses. For objective perfusion analysis ROI measurements on CT/CBCT images were taken in the redistribution-dependent and redistribution-independent liver lobe. Response analysis according to RECIST 1.1 was separately calculated for the redistribution-dependent and redistribution-independent liver lobe. Intrahepatic reperfusion of the embolized variant hepatic artery was observed immediately after embolization with visualization of the subsegmental branches in 95.2% of patients. ROI measurements on CT/CBCT images (right lobe mean 76 ± 30.2 HU, left lobe mean 74.4 ± 30.5, p-value 0.88) did not show any differences. Treatment response after intraarterial chemotherapy did not differ between the redistribution-dependent and redistribution-independent liver lobes. Embolization of hepatic arterial variants during percutaneous arterial port catheter placement results in effective intrahepatic perfusion redistribution and does not compromise treatment efficacy of intraarterial chemotherapy in the redistribution-dependent liver lobe.

A Fast and Accurate Obstacle Segmentation Network for Guava-Harvesting Robot via Exploiting Multi-Level Features

Guava fruit is readily concealed by branches, making it difficult for picking robots to rapidly grip. For the robots to plan collision-free paths, it is crucial to segment branches and fruits. This study investigates a fast and accurate obstacle segmentation network for guava-harvesting robots. At first, to extract feature maps of different levels quickly, Mobilenetv2 is used as a backbone. Afterwards, a feature enhancement module is proposed to fuse multi-level features and recalibrate their channels. On the basis of this, a decoder module is developed, which strengthens the connection between each position in the feature maps using a self-attention network, and outputs a dense segmentation map. Experimental results show that in terms of the mean intersection over union, mean pixel accuracy, and frequency weighted intersection over union, the developed network is 1.83%, 1.60% and 0.43% higher than Mobilenetv2-deeplabv3+, and 3.77%, 2.43% and 1.70% higher than Mobilenetv2-PSPnet; our network achieved an inference speed of 45 frames per second and 35.7 billion floating-point operations per second. To sum up, this network can realize fast and accurate semantic segmentation of obstacles, and provide strong technical and theoretical support for picking robots to avoid obstacles.

Pipeline Optimization for an Economical Microsprinkler Irrigation System Based on a Hybrid Genetic Algorithm

The pipeline combination scheme is optimized to reduce the cost of irrigation units in an economical microsprinkler system. With the standard number of segments of branch and capillary pipes as the decision variables, the lowest pipe cost per unit irrigation area as the objective function, the maximum pressure difference allowed in the system, the minimum number of pipes in the network, and the shape of the irrigation area as constraints, a hybrid genetic algorithm is used to calculate the minimum cost per unit area of the pipe and the corresponding optimal combination of pipelines. The results show that when the irrigation area is a rectangular plot of 30 m × 22 m, the lowest cost of pipeline per unit area, 5192.7 RMB/hm2, is achieved when the capillary pipes are laid bidirectionally along the length of the plot. The cost of bidirectional laying of capillary pipes is less than that of unidirectional laying, and the cost of capillary pipe laid along the length direction of the plot saves 2.3% compared with that of capillary pipe laid along the width direction of the plot. The most extended length of capillary pipe on one side is 103.65 m. Compared with the cost obtained by farmers based on experience, the pipeline cost per hectare was 1.04%∼13.6% lower. It is recommended that farmers lay capillary pipes bidirectionally along the length of the plot and choose branch pipes with smaller diameters to help reduce the cost of the pipe unit area of the irrigation system.

Anatomical study and clinical significance of the posterior ramus of the spinal nerve of the lumbar spine.

Background and objectives: Chronic nonspecific back pain is a common clinical disease typically treated by ultrasound-guided spinal injection. This minimally invasive treatment targets the posterior ramus of the spinal nerve (PRSN). The target of the medial branch is clear, but there is unclear target for the intermediate and lateral branches. This study attempted to observe the distribution of PRSN in the dorsal region of transverse process to provide a more detailed anatomical basis for treating spinal pain. Methods: The present study was conducted on 16 transverse processes of six adult male embalmed corpses. The dorsal area of the transverse process was divided into three equal zones, which are zone I, zone II and zone III from inside to outside. The origin, distribution, quantity, transverse diameter, and relationship with the bone structure of the PRSN on the transverse process were observed. Results: Sixty PRSNs were found in the lumbar of six cadavers, of which 48 were divided into three branches, and 12 PRSNs were divided into two branches. The intermediate branch is mainly distributed in zone I, and the lateral branch is mainly distributed in zone II. Twenty-nine communicating branches were found in 48 adjacent segments of six specimens, all of which originated from the intermediate branch of the previous segment and connected with the lateral branch of the next segment. Conclusion: This anatomical study describing the PRSN may have important clinical significance for spinal surgeons. Understanding the bony localization targets of the PRSN and the links between the PRSNs may benefit patients with low back pain who receive spinal injections.