Search

In deep learning (DL)-based hyperspectral imagery classification, ``spatial patching'' is primarily used as a preprocessing for incorporating local spatial information. This operation can help to promote classification accuracy but it is facing new challenges in the unmanned aerial vehicle (UAV)-borne hyperspectral imagery with high spatial and spectral resolutions (H² imagery). The ground objects' various spatial scales result in it being challenging to determine the optimal size for the spatial patches. In addition, due to the severe spectral variability and spatial heterogeneity of the H² imagery, ``spatial patching'' only exploits the local spatial information and results in serious salt-and-pepper (SP) noise and isolated areas in the classification maps. In this article, to address these issues, a novel spectral patching network (SPNet) with an end-to-end DL architecture is proposed for UAV-borne H² imagery classification. The ``spectral patching'' approach is proposed to preserve the global spatial information and almost all the spectral information of the original hyperspectral imagery. An end-to-end deep encoder-decoder network is then constructed based on the spectral patching mechanism, which introduces the deep residual network (ResNet) and atrous spatial pyramid pooling (ASPP) modules to extract multiscale high-level semantic information for the H² imagery classification. The experimental results obtained with the Wuhan UAV-borne H² imagery (WHU-Hi) UAV-borne hyperspectral data set demonstrate that SPNet can achieve state-of-the-art accuracy and visualization performance in the classification of H² imagery.

BACKGROUND Plexiform angiomyxoid myofibroblastic tumor (PAMT) is a rare mesenchymal tumor characterized by multiple nodular plexiform growth patterns and an immunophenotype with myofibroblasts. The pathological characteristics, immunohistochemistry, diagnostic criteria, differential diagnosis, and gene-level changes of PAMT have been reported in many studies. At present, the main treatment for PAMT in the reported cases is surgery; only eight cases were treated via endoscopy (excluding 1 thoracoscopic resection), and the lesions were all smaller than 5 cm. There are no reports on the prognosis and follow-up of young patients with lesion sizes reaching 5 cm who undergo endoscopic submucosal dissection (ESD). Herein, we present the first case of a young patient with a lesion size reaching 5 cm who was diagnosed with PAMT via endoscopic submucosal dissection. CASE SUMMARY A 15-year-old young man with upper abdominal pain for 2 years presented to the Gastroenterology Department of our hospital. Painless gastroscopy showed a semicircular bulge approximately 5 cm in size in the lesser curvature near the cardia of the fundus; the surface was eroded, and shallow ulcers had formed. The pathological manifestations of the biopsy were spindle cell proliferative lesions with interstitial mucinous changes, and the surface mucosa showed chronic inflammatory changes with active lesions; immunohistochemistry showed smooth muscle actin (SMA) (+), CD117 (-), CD34 (-), DOG-1 (-), S-100 (-), and Ki67 (LI: < 1%). We performed ESD on the patient. The lesion that we removed was 5 cm × 4 cm × 2 cm in size. Pathologically, the resected tissue displayed typical manifestations, such as fat spindle-shaped fibroblasts and myofibroblast-like cells showing irregular nodular hyperplasia. Immunohistochemistry staining of the tumor cells revealed the following: CD34 (partially +), SMA (weakly +), CD117 (-), DOG-1 (-), S-100 (-), SDHB (+), PCK (-), and Ki67 (labelling index: 2%). There was no recurrence or metastasis during the 3-mo follow-up after the operation, and the treatment effect was good. We also performed a review of the literature on the clinical manifestations, pathological features, immunohistochemistry, and differential diagnosis of PAMT. CONCLUSION At present, the diagnostic criteria for PAMT are relatively clear, but the pathogenesis and genetic changes require further study. PAMT is benign in nature, and these patients are less likely to experience local or metastatic recurrence. The main treatment is still surgery if the lesion is in the stomach. Partial gastrectomy and distal gastrectomy are the most frequently performed surgical treatments for PAMT, followed by local resection, subtotal gastrectomy, and wedge resection. But for comprehensive evaluation of the disease, ESD can be considered a suitable method to avoid excessive treatment.

This study seeks to assess the efficacy and safety of transjugular intrahepatic portosystemic shunt (TIPS) combined with transarterial embolization/transarterial chemoembolization (TAE/TACE) in hepatocellular carcinoma (HCC) with portal hypertension and arterioportal shunt (APS). Consecutive hospitalized patients having HCC accompanied by portal hypertension and APS were retrospectively analyzed. A total of 103 patients were enrolled. Of them, 26 patients were in Group A and 77 patients were in Group B according to the treatment protocol (Group A: TIPS plus TAE/TACE; Group B: TAE/TACE alone). The clinical outcomes and survival rate were compared between the two groups. The mean survival time in Group A and Group B were 14 mo and 9.9 mo, respectively, with statistical difference (p = 0.043). The immediate APS improvement rate was 95.2% in Group A and 91.9% in Group B, respectively, with no signficant difference (p = 1.000). However, the first follow-up consultation revealed that APS improvement rate in Group A was more obvious (66.7% vs 27.4%, p = 0.001). Objective response rate of HCC tended to be greater in Group A compared with Group B (65.4% vs 38.7%, p = 0.019). Liver function parameters significantly increased in Group A than those in Group B. After TIPS placement, the mean portal pressure gradient decreased from 32.61 ± 8.87 mmHg to 15.61 ± 8.15 mmHg, with significant difference (p = 0.000). The rate of absorption of ascites and control of variceal bleeding were statistically different between the two groups (p = 0.045 and 0.039, respectively). Our research suggests that TIPS combined with TAE/TACE seems to be safe and efficacious in patients with HCC accompanied by portal hypertension and APS, albeit may be accompanied by liver function damage.

Background Budd-Chiari syndrome (BCS) is a rare heterogeneous liver disease characterized by obstruction of the hepatic venous outflow tract. The incidence of BCS is so low that it is difficult to detect in general practice and difficult to include within the scope of routine diagnosis. The clinical manifestations of BCS are not specific; hence, BCS tends to be misdiagnosed. Case summary We report the case of a 33-year-old Chinese woman who presented with progressive distension in the upper abdomen. She was initially misdiagnosed with liver cirrhosis (LC) due to abnormalities on an upper abdominal computed tomography scan. Although she was taking standard anti-cirrhosis therapy, her symptoms did not improve. Magnetic resonance imaging showed caudate lobe hypertrophy; and dilated lumbar and hemiazygos veins. Venography revealed membranous obstruction of the inferior vena cava owing to congenital vascular malformation. A definitive diagnosis of BCS was made. Balloon angioplasty was performed to recanalize the obstructed inferior vena cava and the patient's symptoms were completely resolved. Conclusion BCS lacks specific clinical features and can eventually lead to LC. Clinicians and radiologists must carefully differentiate BCS from LC. Correct diagnosis and timely treatment are vital to the patient's health.

Evaluating the potential utilization of hybrid energy systems and determining the multi-scale optimal operation strategy is critical to power system planning in the context of energy structure adjustment, especially for large-scale hybrid energy systems. Considering the long-term and short-term complementary characteristics, this paper puts forward a coordinated optimization framework for the integrated energy system in the world’s largest multi-energy complementary base on Yellow River’s upper reaches. The main procedures are as follows: 1) cross-correlation method is introduced for individually analyzing the long- and short-term complementary characteristics of wind power, photovoltaic, and hydropower in this multi-energy complementary base; 2) a double-layer model combining the long-term optimal operation model and short-term optimal operation model for determining the proportion of multiple energy and optimizing the maximum peak-shaving ability; 3) Large-Scale System Decomposition-Coordination Method is applied for solving the proposed double-layer operation model. The results show that wind power 23%, photovoltaic 35%, hydropower 42% can keep the most stable generation in the long-term complementary operation. This proportion results can improve the system peak regulation capacity with 50.8% (sunny day’s morning peak) and 24.2% (rainy day’s morning peak) in the optimal short-term operation.

Since December 2019, COVID-19 has occurred unexpectedly and emerged as a health problem worldwide. Despite the rapidly increasing number of cases in subsequent weeks, the clinical characteristics of pediatric cases are rarely described. A cross-sectional multicenter study was carried out in 10 hospitals across Hubei province. A total of 25 confirmed pediatric cases of COVID-19 were collected. The demographic data, epidemiological history, underlying diseases, clinical manifestations, laboratory and radiological data, treatments, and outcomes were analyzed. Of 25 hospitalized patients with COVID-19, the boy to girl ratio was 1.27:1. The median age was 3 years. COVID-19 cases in children aged

Hepatocellular carcinoma (HCC) is the fourth most lethal malignancy globally. However, most patients present in advanced stages with limited curative treatment options. Therefore, multidisciplinary treatment is often warranted. Here we report a patient with HCC combined with severe arterioportal shunt (APS) who was treated with multidisciplinary approach, comprising interventional radiology procedures, apatinib and camrelizumab. A 31-year-old man who presented to our hospital because of hematemesis caused by HCC with a severe APS. The patient underwent multidisciplinary treatment including transjugular intrahepatic portosystemic shunt, transarterial chemoembolization of arterial branches feeding the shunt and hepatic lesions and subsequent molecular target therapy combined with immunotherapy. After treatment, stable condition of intrahepatic mass and a notable decrease in the number and size of the lung lesions were observed. The patient achieved a long-time survival of more than 2 years. This data suggests that multidisciplinary treatments may be efficacious in the treatment of advanced HCC with severe APS.

PurposeTo investigate the levels of matrix metalloproteinases (MMPs) in aqueous humour of patients with retinal vein occlusion (RVO) and the relationship between intraocular MMP levels and retinal lesion and visual prognosis.Materials and methods52 RVO patients, including 23 with central retinal vein occlusion (CRVO) and 29 with branch retinal vein occlusion (BRVO) and 20 participants with senile cataract were enrolled in this study. Retinal lesions were examined by fundus colour photography, fluorescein fundus angiography and optical coherence tomographic angiography. Sixty microliters of aqueous humour were collected during intravitreal anti-Vascular Endothelial Growth Factor (VEGF) injection or cataract surgery. The aqueous levels of MMP-1, MMP-2, MMP-7, MMP-9 and MMP-10 were measured using the Luminex xMAP multiplex assay. The relationship between MMP levels and clinical presentations was analysed by Pearson correlation test.ResultsThe aqueous humour levels of MMP-1, MMP-2, MMP-7 and MMP-9, but not MMP10 in RVO patients were significantly higher than those in people with cataract after adjusting for age. Further analysis of RVO subgroups showed that the aqueous humour level of MMP2 in CRVO was significantly higher than that in BRVO. The aqueous humour levels of MMP-1 and MMP-2 were positively correlated with superficial capillary plexus vessel density (SVD), whereas the aqueous humour levels of MMP-1 and MMP-7 were negatively correlated with visual improvement following treatment. No correlation between aqueous humour levels of MMP and disease duration and central retinal thickness was observed.ConclusionsRVO eyes had significantly higher intraocular levels of MMP-1, MMP-2, MMP-7 and MMP-9 than cataract eyes and the level of MMP2 appears to be related to the area of occlusion. Intraocular levels of MMP may positively affect SVD and negatively impact visual function in RVO.

The retina is an immune privileged tissue, which is protected from external and internal insults by its blood-retinal barriers and immune suppressive microenvironment. Apart from the avoidance and tolerance strategies, the retina is also protected by its own defense system, i.e., microglia and the complement system. The immune privilege and defense mechanisms work together to maintain retinal homeostasis. During aging, the retina is at an increased risk of developing various degenerative diseases such as age-related macular degeneration, diabetic retinopathy, and glaucomatous retinopathy. Previously, we have shown that aging induces a para-inflammatory response in the retina. In this review, we explore the impact of aging on retinal immune regulation and the connection between homeostatic control of retinal immune privilege and para-inflammation under aging conditions and present a view that may explain why aging puts the retina at risk of developing degenerative diseases.

Radar emitter signal is interfered by various noises during the propagation process, and the signal-to-noise ratio varies widely. Therefore, the features that play a key role in sorting and classifying or identification signals are often difficult to find. In addition, the extracted features are usually subjective and speculative, so it is necessary to select the features that can characterize the maximum difference mode information between the modulated signal categories and the changes in the signal-to-noise ratio. That is, the selected features also have good separability at low SNR. In this paper, based on rough set theory the feature selection method is studied, which lays a foundation for the feature selection of radiation source signals by rough set theory.

Concerned with the problems that the extracted features are the absence of objectivity for radar emitter signal intrapulse data because of relying on priori knowledge, a novel method is proposed. First, this method gets the sparse autoencoder by adding certain restrain to the autoencoder. Second, by optimizing the sparse autoencoder and confirming the training scheme, intrapulse deep features are autoextracted with encoder layer parameters. The method extracts the eigenvectors of six typical radar emitter signals and uses them as inputs to a support vector machine classifier. The experimental results show that the method has higher accuracy in the case of large signal-to-noise ratio. The simulation verifies that the extracted features are feasible.

PURPOSE: The aim of this study was to investigate the aqueous humour levels of elastase-2, lactoferrin, lipocalin-2 (LCN-2), resistin and thrombospondin-1 (TSP-1) in patients with retinal vein occlusion (RVO) and their relationship with visual prognosis following intravitreal anti-vascular endothelial growth factor (VEGF) therapy.MATERIALS AND METHODS: 52 RVO patients (23 cases of central retinal vein occlusion (CRVO) and 29 cases of branch retinal vein occlusion (BRVO)) and 20 cases of senile cataract were enrolled in this study. All RVO patients underwent fundus examinations before and 6-8 months after intravitreal anti-VEGF treatment. Five milliliters of blood were collected from RVO patients before treatment for the measurement of lipids and coagulation factors. Sixty microliters of aqueous humour were collected during intravitreal injection of anti-VEGF or during cataract surgery. The levels of elastase-2, lactoferrin, LCN-2, resistin and TSP-1 in aqueous humour were determined by Luminex xMAP multiple analysis.RESULTS: The aqueous levels of resistin and LCN-2 were significantly higher but the level of TSP-1 was significantly lower in RVO patients compared to controls. Further sub-group analysis showed that CRVO patients had significantly higher levels of resistin and LCN-2 than controls. The aqueous levels of resistin and LCN-2 were negatively correlated with visual improvement following anti-VEGF therapy in CRVO but not in BRVO patients. Visual improvement in RVO patients was not associated with blood lipid levels or any of the coagulation factors.CONCLUSION: CRVO patients had significantly higher aqueous levels of resistin and LCN-2, which negatively impacted on visual improvement after anti-VEGF therapy.

Background: The outbreak of COVID-19 in China was a sudden bio-disaster, which may bring a negative impact on the job burnout of health care professionals (HCPs).Objective: We aim to find out the association factors, especially those closely related to this outbreak, of job burnout in Chinese HCPs.Method: The cross-sectional survey about HCPs' job burnout based on a network platform was conducted in high and low infection regions during the COVID-19 outbreak in China. The demographic characteristics, medical-work-related factors, risk of getting infected due to occupational exposure, and family factors were collected by the self-reported questionnaire. The Chinese version of the Maslach Burnout Inventory (CMBI) and the Trait Coping Style Questionnaire (TCSQ) were employed in this study to evaluate the job burnout and coping style, respectively. Furthermore, statistical analysis was done to find out the associated factors of job burnout.Results: We collected 880 complete questionnaires from doctors and nurses from February 9, 2020 to February 11, 2020. In this study, the positive rates of three dimensions of burnout (emotional exhaustion, depersonalization, and reduced personal accomplishment) and overall burnout were 9.09, 50.57, 56.59, and 73.98%, respectively. After the statistical analysis, we found that several factors can independently affect the dimensions. Working in the high infection region and negative coping styles can affect all three dimensions at once. More night shift quantity and having symptoms could increase emotional exhaustion and depersonalization, while higher work intensity and senior title could increase emotional exhaustion and reduce personal accomplishment, respectively.Conclusion: The rate of moderate and severe burnout had increased due to the outbreak. More attention should be paid to burnout in HCPs, especially those with negative coping. There were some potential ways to reduce burnout, such as reducing their workload and providing better protection from the virus.

The Deep Learning (DL) methods can automatically extract features without artificial prior, provides an effective solution for multi-temporal crop classification. Convolutional Neural Networks (CNNs) have superb spatial-spectral feature extraction capabilities, but often lack consideration of the temporal relationship of multi-temporal images, while Recurrent Neural Networks (RNNs) can better learn the sequential variation pattern. This paper designed a deep spatial-temporal-spectral feature learning network (CropNet) combining the advantages of a deep spatial-spectral feature learning module and a deep temporal-spectral feature learning module for better feature extraction in crop classification from time-series remote sensing images. From the results, the proposed method has better crop classification effects from time-series multispectral images in our experimental areas compared with some common traditional machine learning approaches and common DL methods.

Objectives Artificial intelligence (AI) has been an important addition to medicine. We aimed to explore the use of deep learning (DL) to distinguish benign from malignant lesions with breast ultrasound (BUS). Methods The DL model was trained with BUS nodule data using a standard protocol (1271 malignant nodules, 1053 benign nodules, and 2144 images of the contralateral normal breast). The model was tested with 692 images of 256 breast nodules. We used the accuracy, precision, recall, harmonic mean of recall and precision, and mean average precision as the indices to assess the DL model. We used 100 BUS images to evaluate differences in diagnostic accuracy among the AI system, experts (>25 years of experience), and physicians with varying levels of experience. A receiver operating characteristic curve was generated to evaluate the accuracy for distinguishing between benign and malignant breast nodules. Results The DL model showed 73.3% sensitivity and 94.9% specificity for the diagnosis of benign versus malignant breast nodules (area under the curve, 0.943). No significant difference in diagnostic ability was found between the AI system and the expert group (P = .951), although the physicians with lower levels of experience showed significant differences from the AI and expert groups (P = .01 and .03, respectively). Conclusions Deep learning could distinguish between benign and malignant breast nodules with BUS. On BUS images, DL achieved diagnostic accuracy equivalent to that of expert physicians.

In this paper, a set of wireless communication module based on LoRa technology is developed, and a complete local area network wireless communication solutions satisfying long distance and low power consumption is proposed, which has wide wireless coverage, strong penetration and star architecture. By using LoRa communication, the data of the distribution terminal can be transmitted to the data collection unit seamlessly. The data is more safe and reliable. Through practical field application, the communication system is verified.

Foot ulcers affect 15% of patients with diabetes, resulting in a great health burden. The occurrence and development of diabetic foot ulcers is associated with neuropathy, peripheral arterial disease, and infection. Several growth factors are involved in these processes, including epidermal growth factor, vascular endothelial growth factor, transforming growth factor-beta, fibroblast growth factor, and erythropoietin, which could promote wound healing of patients with diabetes. Thus, this review discusses the role of these growth factors in the pathogenesis of diabetic foot ulcers, aiming to achieve novel insights into the management of diabetic foot ulcers.

Large-scale crop mapping is an important task in agricultural resource monitoring, but it does usually require the ground-truth labels of all the land-cover types in the remotely sensed imagery. However, labeling each land-cover type is time-consuming and labor-intensive. One-class classification, which only needs samples of the class of interest, can solve the problem of redundant labeling. However, the traditional one-class classifiers require well-designed features to realize fine classification, and are thus difficult to apply to complex multi-modal remote sensing data, i.e., optical imagery and synthetic aperture radar (SAR) imagery. In this paper, a deep one-class crop (DOCC) framework that includes a deep one-class crop extraction module and a one-class crop extraction loss module is proposed for large-scale one-class crop mapping. The DOCC framework takes only the samples of one target class as the input to extract the crop of interest by positive and unlabeled learning and can automatically extract the features for one-class crop mapping, without requiring a large amount of labeling for all the land-cover type or feature design based on prior expert knowledge. Experiments conducted on multi-modal remote sensing data, i.e., Zhuhai-1 hyperspectral satellite data, Sentinel-2 multispectral time-series satellite data and Sentinel-1 SAR satellite data, illustrate that DOCC can automatically extract the effective features for one-class classification from multi-modal satellite imagery and reaches the highest F1 accuracy compared with other methods on respective satellite imagery. The results also reveal the different performance of multi-modal satellite imagery when they are used to extract different crop types. Meanwhile, the feasibility of DOCC for multi-modal data can be beneficial for large-scale mapping under different conditions when the samples of multi-class are difficult to obtain.

Crop mapping is essential for agricultural management, economic development planning, and ecological conservation. Remote sensing with a large field of view provides us with a potential technique for large-scale crop mapping. However, most of the previous studies have focused on multi-temporal crop mapping, requiring multiple imaging over a period of time, which is impossible in the cloudy season due to the absence of clear atmospheric windows. Recently, with the progress of spaceborne hyperspectral imaging technology, wide-width hyperspectral satellite images, which provide abundant spectral and spatial structure information, have made the precision crop mapping of large areas possible. This paper focuses on deep learning based crop mapping using one-shot hyperspectral satellite imagery, where three convolutional neural network (CNN) models, i.e., 1D-CNN, 2D-CNN, and 3D-CNN models, are applied for end-to-end crop mapping. In addition, a manifold learning based visualization approach, i.e., t-distributed stochastic neighbor embedding (t-SNE), is introduced to illustrate the discriminative ability of the deep semantic features extracted by the different CNN models. To demonstrate the advantages of one-shot hyperspectral satellite images, an experiment was designed to compare the crop mapping performance of different remote sensing data sources, where both mono-temporal and multi-temporal multispectral images (MSIs) of the same research area were introduced for a systematic comparison. The classification accuracy when using hyperspectral satellite images was found to reach more than 94%, which was much better than that when using mono-temporal MSIs, and was comparable to the result when using multi-temporal MSIs. These findings will be important for the application of hyperspectral data when mapping large-area crop landscapes, and they confirm the potential of CNN models, particularly 3D-CNN models, for crop recognition.

NiTi shape memory alloy (SMA) vascular stent, with good bio-compatibility and super-elastic properties, has become an important medical device in clinical treatment of cardiovascular or cerebrovascular diseases. The free expansion of stent in vessel involves the mechanical behavior of geometric large deformation related to the structure of super-elastic. In this paper, the nonlinear mechanical model for the substructure of NiTi shape memory alloy vascular stent is established, and the displacement deformation equation is derived. The geometric non-linearity of the structure is proved. Furthermore, the geometric nonlinear mechanical model is verified by numerical method. The mechanical analysis model of NiTi shape memory alloy vascular stent is widely used in the field of tracheal stent and so on.

The design of an adaptive fuzzy back-stepping tracking control for flexible air-breathing hypersonic vehicle (FAHV) with actuator constraints is discussed. Fuzzy logic systems (FLSs) are applied to approximate the total uncertainty of FAHV model. The minimal-learning-parameter (MLP) algorithms are applied for FLSs to reduce the quantity of calculation. For the sake of preventing explosion of differentiation term in back-stepping control, the derivatives of virtual control laws can be calculated by sliding mode differentiators. Particularly, the novel auxiliary systems are explored to cope with the inputs saturation. Finally, reference trajectory tracking simulation shows the auxiliary systems have better performances than the existing algorithms when the inputs are saturated.

Deep learning (DL)-based hyperspectral classification primarily use "spatial patching" as preprocessing for incorporating local spatial information. This operation can help to promote classification accuracy but faces the following problems. First, it is difficult to determine the optimal size of spatial patches for different hyperspectral images (HSIs). Second, this operation only exploits spatial features locally but not globally. In this paper, we propose a novel spectral patching network (SPNet) with an end-to-end deep learning architecture for HSI classification. SPNet uses "spectral patching" and Atrous Spatial Pyramid Pooling (ASPP) module to fully preserve the local and global spatial contextual information of original HSIs. The experimental results with UAV-borne hyperspectral dataset demonstrate that the SPNet achieved state-of-the-art accuracy and visualization performance in.

Unmanned aerial vehicle (UAV)-borne hyperspectral systems can acquire hyperspectral imagery with a high spatial resolution (which we refer to here as H2 imagery). As a result of the low operating cost, high flexibility, and the ability to achieve real-time data acquisition, UAV-borne hyperspectral systems have become an important data source for remote sensing based agricultural monitoring. However, precise crop classification based on UAV-borne H2 imagery is a challenging task when faced with a number of different crop classes. The traditional hyperspectral classification methods, such as the spectral-based and object-oriented classification methods, have difficulty in classifying H2 imagery, faced with the problems of salt-and-pepper (SP) noise and scale selection. In this article, the deep convolutional neural network with a conditional random field classifier (CNNCRF) framework is proposed for precise crop classification with UAV-borne H2 imagery. In the proposed method, a deep convolutional neural network (CNN) is designed to extract and fuse in-depth spectral and local spatial features, and the conditional random field (CRF) model further incorporates the spatial-contextual information to improve the problem of holes and isolated regions in the classification map. Meanwhile, virtual sample augmentation based on the hyperspectral imaging mechanism is used to lessen the issue of the limited labeled samples. To validate the results, a new dataset—the Wuhan UAV-borne hyperspectral image (WHU-Hi) dataset—has been built for precise crop classification. The experimental results obtained using the WHU-Hi dataset confirm the accuracy and visualization performance of the proposed CNNCRF classification method, which outperforms the previous methods. In addition, the WHU-Hi dataset could serve as a benchmark dataset for hyperspectral image classification studies.

To date, there has not been a large systematic examination of the hypertrophic cardiomyopathy (HC) phenotype in blacks versus whites. In this study, we investigate differences in presentation of HC between blacks and whites. We included 441 consecutive patients with HC seen at the Johns Hopkins HC clinic in the period from February 2005 to June 2012. We compared 76 blacks for clinical presentation, electrocardiogram, exercise capacity, left ventricular morphology, and hemodynamics by echocardiography to 365 whites. Black patients with HC more often presented with abnormal electrocardiogram (93% vs 80%, p = 0.009), driven by a significant difference in repolarization abnormalities (79% vs 56%, p

Flow characteristics and aerodynamic losses of film cooling injection with compound angles of 30°, 60°, and 90° for velocity ratios of 0.5, 1.0, and 2.0 are numerically investigated using the high-quality hybrid meshes comprising tetrahedrons, prisms, and pyramids. The solutions are obtained by solving the compressible Reynolds-averaged Navier–Stokes equations, and the cell-vertex finite volume method is used to discretize these equations. The predicted results are firstly validated against the experimental data. It has been found that the variation tendency of mass average net total pressure loss coefficient with compound angles is reasonably predicted and in many cases well predicted. The analyses based on net total pressure losses show that the aerodynamic losses tend to increase with the increasing of compound angle regardless of the velocity ratio, and the tendency becomes more obvious when the velocity ratio is large. On the other hand, some typical cases are additionally calculated with different turbulence models and flux computation schemes for the comparisons. A latest proposed numerical flux computation method, simple low-dissipative AUSM (SLAU) scheme, is adopted in the predictions. The results indicate that SLAU scheme is slightly superior to Harten–Lax–van Leer–Einfeldt–Wada scheme in the prediction of compound angle film cooling injection.

Electric upsetting is a complex electrical-thermal-mechanical multi-field coupling technology to the manufacturing process of exhaust valves. Generally, it is difficult to obtain the appropriate forming parameters in the practical industrial production. Thus, finding a convenient and reliable way to choose the optimum process conditions for electric upsetting is a non-trivial issue that still needs to be addressed in greater depth. Based on the multi-field coupling solver platform, MSC.Marc, a finite element model of electric upsetting was established, and then the influence of current and upsetting pressure on deformation were analyzed. It is concluded that as current increases, the maximum temperature and diameter of the deformed material increases, meanwhile as upsetting pressure increases, the maximum temperature decreases, but the maximum diameter of the onion-shape increases. In addition, by means of a three-stage load mode of direct current, the secondary upsetting defect has been avoided. Finally, the finite element model was validated by the results of physical experiments.

Apparent left ventricular cavity dilatation (LVCD) in patients with hypertrophic cardiomyopathy (HCM) is an incompletely understood phenomenon. We aimed at investigating its clinical predictors and potential mechanisms. Sixty one HCM patients underwent N-13-ammonia PET for visual evaluation of LVCD, transient ischemic dilatation (TID) index, myocardial blood flow (MBF), coronary flow reserve (CFR), and regional myocardial perfusion (rMP). TID index was also derived at 2–4 and 15–20 minutes. Visual LVCD and quantitative TID (>1.13 abnormal) agreement were excellent (k 0.91; P

The object of this paper analyzes the reliability level of pre-stressed concrete continuous girder bridges designed with the Chinese codes including the code for design of highway reinforced concrete and pre-stressed concrete bridges and culverts of JTG D62-2004 and JTJ 023-85 using cantilever construction method. Typical cross-sections used in the example bridge are considered at service stage. Load and resistance parameters are treated as random variables. The statistical parameters are based on the available literature, test data and survey results. Reliability indices are calculated by iterations using the first-order second-moment method. The calculated results indicate that the reliability indices on the code of D62-2004 vary considerably to the code of JTJ 023-85. And it could provide the references for the reinforcement of old bridges and the design of new bridges using the cantilever construction.

The design of an adaptive neural back-stepping control for a flexible air-breathing hypersonic vehicle (AHV) in the presence of input constraint and aerodynamic uncertainty is discussed. Based on functional decomposition, the dynamics can be decomposed into the velocity subsystem and the altitude subsystem. To guarantee the exploited controller’s robustness with respect to parametric uncertainties, neural network (NN) is applied to approximate the lumped uncertainty of each subsystem of AHV model. The exceptional contribution is that novel auxiliary systems are introduced to compensate both the tracking errors and desired control laws, based on which the explored controller can still provide effective tracking of velocity and altitude commands when the actuators are saturated. Finally, simulation studies are made to illustrate the effectiveness of the proposed control approach in spite of the flexible effects, system uncertainties, and varying disturbances.

A novel electric-hydraulic hybrid drivetrain incorporating a set of hydraulic systems is proposed for application in a pure electric vehicle. Models of the electric and hydraulic components are constructed. Two control strategies, which are based on two separate rules, are developed; the maximum energy recovery rate strategy adheres to the rule of the maximization of the braking energy recovery rate, while the minimum current impact strategy adheres to the rule of the minimization of the charge current to the battery. The simulation models were established to verify the effects of these two control strategies. An ABS (Anti-lock Braking System) fuzzy control strategy is also developed and simulated. The simulation results demonstrate that the developed control strategy can effectively absorb the braking energy, suppress the current impact, and assure braking safety.

Although the oncogenic role of PPFIA1 (liprin-α1) in breast cancer has been reported, whether its dysregulation is associated with metastasis risk or survival outcomes in breast cancer patients is not clear. Our primary data showed that PPFIA1 expression was significantly higher in liver metastatic breast tumors than in the primary tumors. Then, we tried to pool previous annotated genomic data to assess the prognostic value of PPFIA1 in distant metastasis–free survival, the risk of metastatic relapse, and metastatic relapse–free survival in breast cancer patients by data mining in two large databases, Kaplan–Meier plotter and bc-GenExMiner 4.0. Results from Kaplan–Meier plotter showed that although high PPFIA1 expression was generally associated with decreased distant metastasis–free survival in estrogen receptor+ patients, subgroup analysis only confirmed significant association in estrogen receptor+/N− (nodal negative) group (median survival, high PPFIA1 group vs low PPFIA1 cohort: 191.21 vs 236.22 months; hazard ratio: 2.23, 95% confidence interval: 1.42–3.5, p

Deep convolutional neural networks (CNNs) have been widely used to obtain high-level representation in various computer vision tasks. However, in the field of remote sensing, there are not sufficient images to train a useful deep CNN. Instead, we tend to transfer successful pre-trained deep CNNs to remote sensing tasks. In the transferring process, generalization power of features in pre-trained deep CNNs plays the key role. In this paper, we propose two promising architectures to extract general features from pre-trained deep CNNs for remote scene classification. These two architectures suggest two directions for improvement. First, before the pre-trained deep CNNs, we design a linear PCA network (LPCANet) to synthesize spatial information of remote sensing images in each spectral channel. This design shortens the spatial “distance” of target and source datasets for pre-trained deep CNNs. Second, we introduce quaternion algebra to LPCANet, which further shortens the spectral “distance” between remote sensing images and images used to pre-train deep CNNs. With five well-known pre-trained deep CNNs, experimental results on three independent remote sensing datasets demonstrate that our proposed framework obtains state-of-the-art results without fine-tuning and feature fusing. This paper also provides baseline for transferring fresh pretrained deep CNNs to other remote sensing tasks.

Hypertrophic cardiomyopathy (HCM) is traditionally classified based on a left ventricular outflow tract (LVOT) pressure gradient of 30mmHg at rest or with provocation. There are no data on whether 30mmHg is the most informative cut-off value and whether provoked gradients offer any information regarding outcomes.Resting and provoked peak LVOT pressure gradients were measured by Doppler echocardiography in patients fulfilling guidelines criteria for HCM. A composite clinical outcome including new onset atrial fibrillation, ventricular tachycardia/fibrillation, heart failure, transplantation, and death was examined over a median follow-up period of 2.1years.Among 536 patients, 131 patients had resting LVOT gradients greater than 30mmHg. Subjects with higher resting gradients were older with more cardiovascular events. For provoked gradients, a bi-modal risk distribution was found. Patients with provoked gradients90mmHg (HR 3.92, 95% CI 1.97-7.79) or30mmHg (HR 2.15, 95% CI 1.08-4.29) have more events compared to those with gradients between 30 and 89mmHg in multivariable analysis. The introduction of two cut-off points for provoked gradients allowed HCM to be reclassified into four groups: patients with "benign" latent HCM (provoked gradient 30-89mmHg) had the best prognosis, whereas those with persistent obstructive HCM had the worst outcome.Provoked LVOT pressure gradients offer additional information regarding clinical outcomes in HCM. Applying cut-off points at 30 and 90mmHg to provoked LVOT pressure gradients further classifies HCM patients into low-, intermediate- and high-risk groups.

This paper pertains to a technique of a sensorless indirect stator field-oriented induction motor control, which prevents the accumulative errors incurred by the integrator and the problem relating to the stability of the control system caused by the stator resistance susceptible to temperature variations while conducting the flux estimation directly and computing the synchronous rotary speed. The research adds an adaptive flux observer to estimate the speed of the rotor and uses the fixed trace algorithm (FTA) to execute an online estimation of the slip difference, thereby improving the system of stability under the low rotary speed at regenerating mode and the influence of the rotor resistance on the slip angle. Finally, the paper conducts simulations by Simulink of MATLAB and practices to verify the correctness of the result the paper presents.

Background— Presence of delayed enhancement (DE) on cardiac magnetic resonance (CMR) is associated with worse clinical outcomes in hypertrophic cardiomyopathy. We investigated the relationship between DE on CMR and myocardial ischemia in hypertrophic cardiomyopathy. Methods and Results— Hypertrophic cardiomyopathy patients (n=47) underwent CMR for assessment of DE and vasodilator stress ammonia positron emission tomography to quantify myocardial blood flow and coronary flow reserve. The summed difference score for regional myocardial perfusion was also assessed. Patients in the DE group (n=35) had greater left ventricular wall thickness (2.09±0.44 versus 1.78±0.34 cm; P =0.03). Stress myocardial blood flow (2.25±0.46 versus 1.78±0.43 mL/min per gram; P =0.01) and coronary flow reserve (2.78±0.32 versus 2.01±0.52; P P P Conclusions— We found a close relationship between DE by CMR and microvascular function in most of the patients studied. However, a small proportion of patients had DE in the absence of perfusion abnormalities, suggesting that microvascular dysfunction and ischemia are not the sole causes of DE in hypertrophic cardiomyopathy patients.

A speed estimation approach of vector-controlled induction motor drive based on flux observer that applying particle swarm optimization (PSO) algorithm is presented. The proposed estimation speed scheme using full-order stator flux observer has the advantage of simple structure and facile implementation. The observer gain matrix of the full-order flux observer is designed applying the particle swarm optimization algorithm which acquired exact gain parameters rapidly. Simulation and experimental results confirms the effectiveness of the proposed approach.

The sensorless field oriented technique has a very broad application and is often applied to motor drive control systems to enhance the operating accuracy and efficiency. However, this technology needs to design a set of estimator which can estimate actual values of the stator resistance, the rotor resistance and the rotor speed to prevent the occurrence of the unstable drive system caused by matching parameters subject to temperature or other ambient factors. There is an implicit problem when multiple estimators work at the same time. The problem is that the dependency between the estimators will fail to estimate. The strategies for estimation presented in references often ignore the dependency between the estimators. Thus, estimation errors are incurred, and the quality of the drive system is affected. This study submits a set of speed sensorless induction motor drive system which adds a flux observer to redeem the operating stability at low rotary speeds, builds an authentication theorem forthe dependency between multiple estimators which are coupled to each other according to Popov's criterion, and takes advantage of the neoteric Particle Swarm Optimization (PSO) algorithm to adjust the character of the relation between stator and rotor resistance estimators. Finally, this study takes an indirect rotor-field-orientated control as a testing architecture and simulates the speed sensorless control technique by MATLAB Simulink® to prove the feasibility of this control system.

An approach of air-to-air autonomous attack occupation for Unmanned Combat Aerial Vehicles (UCAVs) is proposed to improve attack precision and combat effectiveness. According to the shortage of UCAV in the task of attack occupation, kinematic and dynamic models of UCAV and missile loaded on it are formed. Then, attack zone and no-escape zone are calculated by pattern search algorithm, and the optimum attack position is indicated. To arrive at the optimum attack position accurately with restriction of gesture, a novel adaptive sliding mode control method is suggested to design the integrated guidance and control system of UCAV in the process of autonomous attack occupation. Key parameters of the control system are adaptively regulated, which further economize control energy at the same time. The simulation results show that compared with traditional methods our approach can guide the UCAV to the optimum attack position with stable gesture and economize nearly 25% control energy.

An automated multi-objective and multidisciplinary design optimization (MDO) of a transonic turbine stage to maximize the isentropic efficiency and minimize the maximum stress of the rotor with constraints on mass flowrate and dynamic frequencies is presented in this article. The self-adaptive multi-objective differential evolution (SMODE) algorithm is studied and developed to seek Pareto solutions of the optimization, and a new constraint-handling method based on multi-objective optimization concept is applied for constraint handling. The optimization performance of the presented SMODE is demonstrated using the typical mathematical tests. By applying SMODE as an optimizer and integrating three-dimensional (3D) blade modelling method based on non-uniform B-spline, load-fitting transfer algorithm in parameter space, 3D Navier–Stokes solution technique, and finite element analysis method as well, seven Pareto solutions are obtained. Two Pareto solutions are analysed in detail. One is the highest isentropic efficiency individual, while the other is a compromise between efficiency and mechanical stress in the blade. The aerodynamic performance and strength characteristics of the optimized turbine stage are significantly improved. The analysis results indicate that the presented multi-objective and MDO method has a potential in the optimization of blade performance and can be applied as a promising method for the optimization design of axial turbomachinery blades

Astrocyte elevated gene-1 (AEG-1) plays an important role in tumor progression including transformation, evasion of apoptosis, invasion, metastasis, and chemoresistance. However, the expression of AEG-1 in ovarian carcinoma and its significance are still unclear. The aim of this study was to examine the expression of AEG-1 in ovarian carcinoma. Immunohistochemistry was performed to determine the expression of AEG-1 in 81 ovarian carcinoma specimens. The correlation of AEG-1 expression with clinicopathological parameters was assessed using χ 2 analysis. Patient survival was analyzed using the Kaplan-Meier and log-rank tests. Cox regression was used for the multivariate analysis of prognostic factors. High expression of AEG-1 was detected in 66.67% of the ovarian carcinomas and was significantly associated with the International Federation of Gynecology and Obstetrics stage, histological grade, presence of residual tumor after primary surgery, and tumor recurrence. Patients with high AEG-1 expression had significantly poorer overall survival and disease-free survival (both P

To reduce the ladder axis machining error, in this paper, the finish turning of a machine tool spindle and the rough turning process is analyzed, were calculated the change of step-axis contour deformation in the process of the finish turning and rough turning with limited difference method and the finite element method (the elastic constraint and inelastic constraint in both ends ). Comparison of the deformation of the three profile, to find more realistic results, and simulate the spline of shaft ladder deformation, using Compensation method to process NC programming. To improve the processing quality of stepped shaft.

BACKGROUND: Lysosome-associated protein transmembrane 4 beta (LAPTM4B), a novel oncoprotein, is overexpressed in several carcinomas. Previous studies revealed that LAPTM4B polymorphisms contribute to the risk of certain types of cancers. The purpose of this study was to investigate the association between different LAPTM4B alleles and the risk of cervical carcinoma. METHODS: A case-control analysis was performed in 317 patients with cervical cancer and 416 control subjects. Genomic DNA was extracted from peripheral blood lymphocytes in all participants. LAPTM4B genotypes were determined using polymerase chain reaction. Odds ratios and 95% confidence intervals (CIs) were computed using an unconditional logistic regression model. RESULTS: There was a significant difference (P

Advanced hepatocellular carcinoma (HCC) with tumor thrombi invading the portal vein and extending into the right atrium (RA) through the hepatic vein is regarded as a terminal-stage condition. Intracardiac tumor thrombus and treatment via liver resection has been reported in the current literature, but results from this therapeutic approach remain unsatisfactory. The present study describes a rare case of HCC with metastatic portal vein, middle hepatic vein, inferior vena cava (IVC) and RA tumor thrombi, and pulmonary metastases. A 29-year-old woman was admitted to The First Affiliated Hospital of Guangxi Traditional Chinese Medical University (Nanning, China) subsequent to experiencing right upper quadrant abdominal pain. Following diagnosis, based on computed tomography analysis and laboratory data, the patient underwent an initial transcatheter arterial chemoembolization (TACE) treatment using fluorouracil (5-FU), pirarubicin, mitomycin C, Lipiodol and sodium alginate microball (KMG). At 1 month post-treatment, serum α-fetoprotein levels remained at >1,000 ng/ml. Subsequently, the patient underwent a second TACE treatment. At 1 month after the second treatment, the abdominal pain had been alleviated and the serum α-fetoprotein levels were reduced to

Aims To investigate whether hypertrophic cardiomyopathy (HCM) patients with late gadolinium enhancement (LGE) confined to the right ventricular insertion points (RVIP) differ phenotypically from patients without LGE or intramural LGE in the left ventricle (LV). Methods and results Sixty-two HCM patients underwent cardiac magnetic resonance for quantification of LGE (% LV mass) and were classified as group (i) no-LGE ( n = 18), group (ii) LGE-RVIP ( n = 19), and group (iii) intramural LGE ( n = 25). All patients also underwent vasodilator N-13 ammonia PET to quantify myocardial blood flow (MBF) and myocardial flow reserve (MFR), and echocardiography to measure longitudinal LV strain. LGE extent (17 ± 11% vs. 4 ± 4% vs. 0%; P < 0.001) and LV thickness (21.7 ± 3.4 vs. 18.8 ± 3.9 vs. 16.3 ± 2.8 mm; P < 0.001) were significantly greater in group 3, intermediate in group 2, and lower in group 1. In contrast, stress MBF (1.62 ± 0.44 vs. 1.90 ± 0.37 vs. 2.22 ± 0.48 mL/min/g; P < 0.001); MFR (1.92 ± 0.47 vs. 2.15 ± 0.52 vs. 2.71 ± 0.52; P < 0.001), and longitudinal LV strain (−11.4 ± 3.8 vs. −12.6 ± 3.2 vs. −14.4 ± 4.1%; P = 0.04) were lower in group 3, intermediate in group 2, and higher in group 1. Conclusions From an imaging viewpoint, patients with LGE confined to only the RVIP appear to represent an intermediate-stage phenotype between patients with no LGE and intramural LGE in the LV.

Deep convolutional neural networks (CNNs) have been widely used to obtain high-level representation in various computer vision tasks. However, for remote scene classification, there are not sufficient images to train a very deep CNN from scratch. From two viewpoints of generalization power, we propose two promising kinds of deep CNNs for remote scenes and try to find whether deep CNNs need to be deep for remote scene classification. First, we transfer successful pretrained deep CNNs to remote scenes based on the theory that depth of CNNs brings the generalization power by learning available hypothesis for finite data samples. Second, according to the opposite viewpoint that generalization power of deep CNNs comes from massive memorization and shallow CNNs with enough neural nodes have perfect finite sample expressivity, we design a lightweight deep CNN (LDCNN) for remote scene classification. With five well-known pretrained deep CNNs, experimental results on two independent remote-sensing datasets demonstrate that transferred deep CNNs can achieve state-of-the-art results in an unsupervised setting. However, because of its shallow architecture, LDCNN cannot obtain satisfactory performance, regardless of whether in an unsupervised, semisupervised, or supervised setting. CNNs really need depth to obtain general features for remote scenes. This paper also provides baseline for applying deep CNNs to other remote sensing tasks.