Search

Postpartum physical pain is common in clinic, mostly manifested in pain, numbness or weight, swelling and so on of limbs and joints during puerperium, and some of them fail to heal over the years, becoming stubborn “puerperal fever.” Mr. Zhang has been a doctor for more than 30 years. He has superb medical skills and has unique opinions on postpartum diseases. The author is lucky to follow-up. Now, his experience in syndrome differentiation and treatment of postpartum body pain is analyzed as follows, and the tested case is attached.

We conducted in this paper a regression analysis of factors associated with acute radiation pneumonia due to radiation therapy for lung cancer utilizing cluster analysis to explore the predictive effects of clinical and dosimetry factors on grade ≥2 radiation pneumonia due to radiation therapy for lung cancer and to further refine the effect of the ratio of the volume of the primary foci to the volume of the lung lobes in which they are located on radiation pneumonia, to refine the factors that are clinically effective in predicting the occurrence of grade ≥2 radiation pneumonia. This will provide a basis for better guiding lung cancer radiation therapy, reducing the occurrence of grade ≥2 radiation pneumonia, and improving the safety of radiotherapy. Based on the characteristics of the selected surveillance data, the experimental simulation of the factors of acute radiation pneumonia due to lung cancer radiation therapy was performed based on three signal detection methods using fuzzy mean clustering algorithm with drug names as the target and adverse drug reactions as the characteristics, and the drugs were classified into three categories. The method was then designed and used to determine the classification correctness evaluation function as the best signal detection method. The factor classification and risk feature identification of acute radiation pneumonia due to radiation therapy for lung cancer based on ADR were achieved by using cluster analysis and feature extraction techniques, which provided a referenceable method for establishing the factor classification mechanism of acute radiation pneumonia due to radiation therapy for lung cancer and a new idea for reuse of ADR surveillance report data resources.

Objective This study was aimed to determine how platelet reactivity (PR) on dual antiplatelet therapy predicts ischemic and bleeding events in patients underwent percutaneous coronary intervention (PCI). Design A total of 2768 patients who had received coronary stent implantation and had taken aspirin 100 mg in combination with clopidogrel 75 mg daily for > 5 days were consecutively screened and 1885 were enrolled. The recruited patients were followed-up for 12 months. The primary end-point was the net adverse clinical events (NACE) of cardiovascular death, nonfatal myocardial infarction (MI), target vessel revascularization (TVR), stent thrombosis (ST) and any bleeding. Result 1709 patients completed the clinical follow-up. By using the receiver operating characteristic (ROC) curve analysis, the optimal cut-off values were found to be 37.5 and 25.5% respectively in predicting ischemic and bleeding events. Patients were classified into 2 groups according to PR: inside the window group (IW) [adenosine diphosphate (ADP) induced platelet aggregation (PLADP) 25.5–37.4%)] and outside the window group (OW) (PLADP P = 0.004)] than that in the OW group during 12-month follow-up. Conclusion An optimal therapeutic window of 25.5–37.4% for PLADP predicts the lowest risk of NACE, which could be referred for tailored antiplatelet treatment while using LTA assay. Trial registration Trial registration number: ClinicalTrials.govNCT01968499. Registered 18 October 2013 - Retrospectively registered.

The main objective of this study is to analyze the clinical and pathological features and prognosis of patients with Hepatitis B associated membranous nephropathy (HBV-MN) and idiopathic membranous nephropathy (IMN) complicated with hepatitis B virus (HBV) infection. This study will provide more basis for diagnosis and prognosis evaluation. A total of 50 patients with HBV-MN were included in this study. 56 IMN patients complicated with HBV infection diagnosed during the same period formed the control group. Parameters including blood routine, urine routine and plasma levels of albumin (ALB), serum creatinine (SCR), blood urea nitrogen (BUN), urea acid (UA), total cholesterol (T-CHO), triglycerides (TG), complement C3 and C4, glutamic pyruvic transaminase (ALT), glutamic pyruvic transaminase (AST), 24-h urinary protein quantification (24 h-TP), renal phospholipase A2 receptor (PLA2R) and HBV related markers during the hospitalization and outpatient follow-up study period were collected for all the patients. The proportion of male patients was high in both groups. The average age of the HBV-MN group was 37.2 ± 14.187 years old, it was younger compared with the IMN group (P = 0.003). Nephrotic syndrome was the major clinical manifestation among patients. There was no significant difference between the two groups in the levels of anemia, microscopic hematuria, renal dysfunction, liver dysfunction, liver cirrhosis. The level of serum C3 and C4 in the HBV-MN group was lower compared with the IMN group (P = 0.002, P = 0.014). In the HBV-MN group, serum HBV markers were negative in 6 (12%) patients, 4 patients (8%) were positive for PLA2R in serum, and 5 patients (10%) were positive for PLA2R in renal tissue. Stronger IgG1 and C1q and weaker IgG4 staining were found in HBV-MN group renal tissues (P = 0.003, P = 0.025, and P = 0.001, respectively). There were no statistical differences compared with serum and renal PLA2R between HBV-MN and IMN groups (P = 0.098, P = 0.109). During the 1-year follow-up, there was no significant difference in complete remission rate between the two groups (P = 0.7739). Renal biopsy is crucial to diagnose HBV-MN. IgG subtypes in the HBV-MN group were mainly IgG1 deposition, while those in IMN complicated with HBV infection group were mainly IgG4 deposition. When HBV-associated antigen and PLA2R are present in renal tissue, lower level of serum C3 and C4, high intensity of renal C1q and IgG1 is more supportive of HBV-MN. The positive of PLA2R in serum and renal tissue in differentiating HBV from IMN complicated with HBV infection remains to be discussed.

Image segmentation is a basic problem in medical image analysis and useful for disease diagnosis. However, the complexity of medical images makes image segmentation difficult. In recent decades, fuzzy clustering algorithms have been preferred due to their simplicity and efficiency. However, they are sensitive to noise. To solve this problem, many algorithms using non-local information have been proposed, which perform well but are inefficient. This paper proposes an improved fuzzy clustering algorithm utilizing nonlocal self-similarity and a low-rank prior for image segmentation. Firstly, cluster centers are initialized based on peak detection. Then, a pixel correlation model between corresponding pixels is constructed, and similar pixel sets are retrieved. To improve efficiency and robustness, the proposed algorithm uses a novel objective function combining non-local information and a low-rank prior. Experiments on synthetic images and medical images illustrate that the algorithm can improve efficiency greatly while achieving satisfactory results.

Objectives: By analyzing the protective effect and mechanism of tobacco on knee joint cartilage in rats, this paper studies the effect of tobacco on knee joint recovery of college students after sports. Methods: Firstly, the main subunits of nAChRs were systematically studied by using the rat knee arthritis model α 7 and α 4 and β To clarify the correlation between nAChRs and the occurrence and development of OA. Then, the OA rat model prepared by iodoacetic acid was used as the experimental object to observe the protective effect of nicotine on knee osteoarthritis cartilage in rats. Results: The histological changes of rats in MIA group were obvious after operation. The results of light microscope score and Mankin's score at 15 and 30 days were significantly higher than those in con group. Of right knee cartilage in rats in MIA group α 7, α 4 and β The expression of 2 did not change significantly on the 15th day, but increased significantly on the 30th day compared with the blank control group. Conclusion: Nicotine has a protective effect on knee bone and joint cartilage and promotes the accelerated recovery of knee bone and joint after exercise.. Key words: nicotine, knee joint, cartilage, recovery after exercise.

Prussian blue (PB) is a well-known electrochromic material due to its fast response and high cycle stability. We used the multi-step electrodeposition to prepare the PB films (S-PB) with different thickness. SEM image shows that as-grown film has small particles on its surface, which is beneficial for improving electrochromic properties of thicker PB film. The S-PB film with a thickness of 410 nm shows a high transmittance modulation of 82.04% at λ = 680 nm. S-PB and CeO2-TiO2-Li films were used as working electrode and counter electrode separately to prepare electrochromic devices (ECDs), both of which are inorganic anode materials. At 680 nm wavelength, the device based on double anode materials changes transmittance from 71.25 to 11.71% and the coloration efficiency is 73.5 cm2/C. Moreover, the optical modulation of ECD does not change significantly when performing 10,000 cycles.

It is conventional to assume that the role of the soil-structure interaction (SSI) is beneficial to the buildings under seismic loading. However, lessons learned from recent earthquakes revealed that this assumption could be misleading, and SSI may have different effects on the seismic response of different structural systems. In this study, an enhanced soil-structure numerical model is developed and verified using ABAQUS software to assess the impact of SSI on high-rise frame-core tube structures. The seismic responses of 20, 30, and 40-storey buildings constructed on soil class Ee (according to Australian Standards) under four earthquake acceleration records have been studied. The results in terms of maximum lateral deflections, foundation rocking, inter-storey drifts and storey shear forces for the rigid base and flexible base frame-core tube structures have been discussed and compared. Generally, SSI has a remarkable impact on the seismic behaviour of high-rise frame-core tube structures since it can increase the lateral deflections and inter-storey drifts and decrease storey shear forces of structures. However, It is worth noting that the seismic responses of soil-structure systems under near and far field earthquakes are considerably different.

Biomass fuel has been widely concerned because its net CO2 emission is close to zero. Biomass boilers are known to have lower pollutant emissions than fossil fuel boilers, but in some applications, they also release high-level CO and NO. We developed a medium-sized hydrogen and oxygen (HHO) generator, with high energy conversion rate and adjustable output gas. The HHO gas was then introduced into a biomass hot air generator for mixed combustion. The experimental results showed that based on the electricity consumption of gas production and biomass fuel price, the total cost during preheating reduced. In addition, the average concentrations of CO, NO and smoke decreased by 93.0%, 22.5% and 80%, respectively. Integration of biomass fuel and HHO gas can effectively reduce pollutant emissions and save fuel, especially in areas rich in renewable energy.

Background Postoperative radiation is one of the most effective treatments for keloid. Radiation-induced hyperpigmentation is, however, a concern for both patients and dermatologists. Aims To identify clinical features and risk factors of superficial X-ray-induced hyperpigmentation in postoperative keloid radiotherapy. Patients/methods From November 2019 to November 2020, 70 keloids in 55 patients were treated with superficial X-ray after surgery. Patient data and treatment-related parameters were collected. Follow-up photographs were taken one month after radiotherapy. Three independent dermatologists assessed photographs for pigmentation. The clinical characteristics and severity of pigmentation were summarized. Logistic regression was used to analyze the risk factors associated with occurrence and severity of hyperpigmentation. Results The pigmentation was a well-defined rectangular brown macule consistent with the unprotected area that involves normal skin, and 82.9% of keloids displayed pigmentation induced by superficial X-ray at one month after treatment. All keloids of trunk appeared to be mild-to-moderately hyperpigmented (100%); 92.89% of limb keloids displayed mainly severe pigmentation. Keloid location was an independent risk factor for the incidence and severity of pigmentation. Longer incision length was also predictive of increased hyperpigmentation severity. Conclusions Superficial X-ray-induced hyperpigmentation was common and severe in postoperative keloid radiotherapy. The incidence and severity of pigmentation can be predicted by the location and incision length of the keloid.

Background Recruitment of volunteers is a major challenge for clinical trials. There has been increasing development and use of Internet-based portals in recruitment for clinical research. There has been little research on researcher use and perceptions of these portals. Objectives This study evaluated researcher perceptions of use of Rally, an Internet-based portal for clinical trial volunteer recruitment. Methods A cross-sectional survey was developed and implemented to understand researcher perceptions. From theoretical models of information technology use, the survey adopted items in four domains: ease of use, usefulness, facilitating conditions, and self-efficacy. The dependent variable was researchers' behavioral intention to use Rally. The survey captured characteristics of researchers such as gender, age, and role. It was implemented using the REDCap survey tool. An email invitation followed by three reminders was sent to researchers. A hierarchical regression model was applied to assess predictors of behavioral intention. Results The survey response rate was 35.6% (152 surveys received from 427 contacted researchers). In the hierarchical regression model, facilitating conditions and self-efficacy predicted behavioral intention (F (4,94) = 6.478; p Conclusion Facilitating conditions and self-efficacy predicted researchers' behavioral intention to use Rally for volunteer recruitment into clinical trials. Future research should document best practices and strategies for enhancing researcher use of online portals for volunteer recruitment.

Image inpainting is an essential issue in the field of computer vision. The basic purpose is to automatically recover the lost content according to the known content in the image. Although signific...

Objective. To evaluate the diagnostic value of gastrin-17 (G-17) and pepsinogen (PG) in gastric cancer (GC) screening in China, especially eastern China, and to determine the best diagnostic combination and threshold (cutoff values) to screen out patients who need gastroscopy. Methods. The serum concentrations of G-17 and pepsinogen I and II (PGI and PGII) in 834 patients were analyzed, and the PGI/PGII ratio (PGR) was calculated. According to pathological results, patients can be divided into chronic nonatrophic gastritis (NAG)/chronic atrophic gastritis (CAG)/intraepithelial neoplasia (IN)/GC groups. The differences in G-17, PG, and PGR in each group were analyzed, and their values in GC diagnosis were evaluated separately and in combination. Results. There were differences in serum G-17, PGII, and PGR among the four groups (NAG/CAG/IN/GC) ( P ≤ 0.001 ). In total, 54 GC cases were diagnosed, of which 50% were early GC. There was no significant difference in the PGI levels among the four groups ( P = 0.377 ). NAG and CAG composed the chronic gastritis (CG) group. The G-17 and PGII levels in the IN and GC groups were higher than those in the CG group (both P ≤ oth C ), while the PGR levels were lower ( P ≤ lower ). When distinguishing NAG from CAG, the best cutoff value for G-17 was 9.25 pmol/L, PGII was 7.06 μg/L, and PGR was 12.07. When distinguishing CG from IN, the best cutoff value for G-17 was 3.86 pmol/L, PGII was 11.92 μg/L, and PGR was 8.26. When distinguishing CG from GC, the best cutoff value for G-17 was 3.89 pmol/L, PGII was 9.16 μg/L, and PGR was 14.14. The sensitivity, specificity, accuracy, and positive and negative predictive values of G-17/PGII/PGR for GC diagnosis were 83.3%/70.4%/79.6%, 51.8%/56.3%/47.8%, 53.8%/57.2%/49.9%, 10.7%/10.9%/9.6%, and 97.8%/96.5%/97.1%, respectively. The sensitivity, specificity, accuracy, and positive predictive and negative predictive values of PGII/G-17 vs. PGR/G-17 vs. PGR/PGII in the diagnosis of GC were 63.0% vs. 70.4% vs. 64.8%, 70.5% vs. 70.1% vs. 60.4%, 70.0% vs. 70.1% vs. 60.7%, 12.9% vs. 14.0% vs. 10.2%, and 96.5% vs. 97.2% vs. 96.1%, respectively. Conclusion. The PGII and G-17 levels in patients with gastric IN and GC were significantly increased, while the serum PGR level was significantly decreased. Serological detection is effective for screening GC. The combination of different markers can improve the diagnostic efficiency. The highest diagnostic accuracy was G-17 combined with PGR, and the best cutoff values were G − 17 > 3.89 pmol / L and PGR < 14.14 .

The overuse of agrochemicals for agricultural productivity to meet the global food demand of the rapidly growing human population is a great environmental threat, particularly for aquatic ecosystems. Being associated intimately with plant health, growth, and productivity, the plant microbiome is emerging as a promising environmentally friendly and sustainable resource for agricultural productivity. For the past decades, our understanding of the interactions between plants and microorganisms and our knowledge of how to improve the plant microbiome by using microbial inoculants has increased significantly. A better understanding of the impact of the plant microbiome on mineral resources will benefit plant and soil health. In this review, we highlight the importance of microbial inoculants and their interactions with mineral fertilizers in enhancing crop productivity, as well as current challenges.

This paper proposes a fully digital signal processing scheme for ultrasonic Doppler endoscope imaging. 200 patients with superficial tissue lumps were randomly divided into two groups: the control group and the experimental group. These two groups used conventional ultrasound examination and colored Doppler ultrasound imaging technology, respectively, to observe and compare the test method and the surgical pathological examination results. Compared with the results of the two groups, the diagnostic compliance rate of the patients in the experimental group was 99.0% significantly higher than the diagnostic compliance rate of 86.0%. At the same time, 300 patients with surgery and pathologically confirmed superficial organ lesions were selected in the hospital, and all patients were diagnosed by ultrasound to observe the diagnosis. The clinical effects of colored Doppler ultrasound to diagnose vascular lesions in the lower extremities of diabetes were discussed. The rate of arteriosclerosis in the lower extremities of the observation group was 92.32%. The more than 50% vascular stenosis rate was 45.16%. The vascular blocking rate was 16. 13% and thrombosis rate 6.45% were significantly higher than the control group 12.90%, 8.06%, 0.00%, 0.00%. In the diagnosis of superficial tissue lesions, the resolution of colored Doppler ultrasound imaging technology is relatively high, which can significantly improve the clinical diagnosis rate and has clinical application value.

Accurate image segmentation is a prerequisite to conducting an image analysis task, and the complexity stemming from the semantic diversity plays a pivotal role in image segmentation. Existing algorithms employed different types of information in the process of segmentation to improve the robustness. However, these algorithms were characterized by a tradeoff between noise removal and detail retention; this is because it is difficult to distinguish image artifacts from details. This paper proposes an improved image segmentation schema and presents two improved clustering algorithms, in which self-similarity and back projection are considered simultaneously to enhance the robustness. With the aid of self-similarity, non-local information is fully exploited, while the original information can be retained by back projection. Extensive experiments on various types of images demonstrate that our algorithms can balance noise restraining and detail retention to improve the adaptation of complex images in segmentation.

The optimal surgical modality for duodenal gastrointestinal stromal tumor (GIST) remains undefined. The purpose of this study was to evaluate long-term survival outcomes of patients who underwent radical resection (RR) or limited resection (LR) of duodenal GIST. A total of 325 patients identified from the Surveillance, Epidemiology and End Results (SEER) database who underwent surgery for duodenal GIST between 1986 and 2016 were classified into a LR group and a RR group based on the type of surgery received. Propensity score matching (PSM) was performed to minimize the selection bias in comparisons. Disease-specific survival (DSS) and overall survival (OS) were observed, and factors affecting the survival outcome were analyzed. In the entire cohort, 105 patients (32.3%) underwent RR and 220 (67.7%) received LR. Both the 5-year OS and DSS in RR group were significantly better than those in LR group (71.0% vs. 54.1%, P = 0.014; 66.6% vs. 49.1%, P = 0.025). PSM resulted in 95 pairs of patients, with long-term outcomes being comparable between the two groups. After adjusting covariates in the propensity matched cohort, the type of surgery still showed no significant impact on OS (hazard ratio [HR] 1.160; 95% confidence interval [CI] 0.662–2.033) and DSS (HR 1.208; 95% CI 0.686–2.128). Surgical modalities do not seem to have a significant impact on long-term survival outcomes of patients with duodenal GIST and should mainly depend on the tumor size and location.

In this study, all-fiber optic current sensor is adapted to measure the leakage current of ships. The analytical model is built in a two-dimensional (2D) cartesian coordinate system. The magnetic field distribution of the weak current of a single long straight wire and the leakage current formed by two wires are modeled and simulated, respectively. In addition, the correctness of the analytical model is verified with the finite element method. An investigation is conducted on the conditions of the formula application of the all-fiber current sensor. According to the result, no magnetic field mutation, sensing fiber coil closure, transmission fiber insensitive to the magnetic field, and stable operation of the Verdet constant are demonstrated as the prerequisites for the combination of the Faraday effect and Ampere loop theorem. Lastly, the experiments are performed to verify the feasibility of the method proposed in this study. The sensor exhibits an accuracy of 1% and high linearity when the leakage current ranges from 1mA to 99mA.

Deep learning has revolutionized the performance of classification and object detection, but meanwhile demands sufficient labeled data for training. Given insufficient data, while many techniques have been developed to help combat overfitting, the challenge remains if one tries to train deep networks, especially in the ill-posed extremely low data regimes : only a small set of labeled data are available, and nothing – including unlabeled data – else. Such regimes arise from practical situations where not only data labeling but also data collection itself is expensive. We propose a deep adversarial data augmentation (DADA) technique to address the problem, in which we elaborately formulate data augmentation as a problem of training a class-conditional and supervised generative adversarial network (GAN). Specifically, a new discriminator loss is proposed to fit the goal of data augmentation, through which both real and augmented samples are enforced to contribute to and be consistent in finding the decision boundaries. Tailored training techniques are developed accordingly. To quantitatively validate its effectiveness, we first perform extensive simulations to show that DADA substantially outperforms both traditional data augmentation and a few GAN-based options. We then extend experiments to three real-world small labeled classification datasets where existing data augmentation and/or transfer learning strategies are either less effective or infeasible. We also demonstrate that DADA to can be extended to the detection task. We improve the pedestrian synthesis work by substitute for our discriminator and training scheme. Validation experiment shows that DADA can improve the detection mean average precision (mAP) compared with some traditional data augmentation techniques in object detection. Source code is available at https://github.com/SchafferZhang/DADA .

Dielectric capacitors, which have the characteristics of greater power density, have received extensive research attention due to their application prospects in pulsed power devices. Film capacitors are easier to integrate into circuits due to their smaller size and higher energy storage density compared to other dielectric capacitor devices. Recently, film capacitors have achieved excellent energy storage performance through a variety of methods and the preparation of multilayer films has become the main way to improve its energy storage performance. In this review, we have summarized several control optimization mechanisms, such as heterojunction effect, interfacial ‘dead-layer’ and space-charges effect, modulating the distribution of electric field and polarization, multilayer film interface design and interface blocking effects, and interlayer coupling, for multilayer films by recalling previous studies on multilayer films. The outlook and future research direction of multilayer films are also discussed in this article.

Image deblurring aims to restore the latent sharp image from the blurred one. In recent years, some learning-based image deblurring methods have achieved significant advances. However, the tradeoff between the texture details and model parameters is still a crucial issue. In this paper, we propose a novel deblurring method based on two-level wavelet-based convolutional neural network (CNN), which embeds discrete wavelet transform (DWT) to separate the image context and texture information and reduces the complexity of calculation. Furthermore, we modify the Inception module by adding pixels-wise attention (PA) mechanism and channel scaling factor to make each convolution kernel have different weights, which increase the receptive field while significantly reduce the parameters of the module. Qualitative and quantitative evaluation on real-word and synthetic datasets shows that the deblurring performance of our method is comparable to the state-of-the-art algorithms. Moreover, compared to the traditional learning-based deblurring method, our model has fewer parameters.

BACKGROUND Inflammation and postinflammatory hyperpigmentation (PIH) are two adverse side effects of ablative fractional CO2 laser (ablative Fr CO2 ) treatment for both rejuvenation and acne scars. AIMS To evaluate the efficacy of fusidic acid cream versus erythromycin ointment as postoperative therapy in patients who underwent ablative Fr CO2 treatment for atrophic acne scars. METHODS Sixty Chinese patients who fulfilled the study criteria were recruited and randomly assigned to an experimental group treated with fusidic acid cream or a control group treated with erythromycin ointment. Postoperative treatment lasted for 7 days in each group. Each patient had two follow-up visits at 8 and 12 weeks. In total, three patients dropped out of the study, one in the experimental group at week 8 and two in the control group at weeks 8 and 12. Baseline and postoperative inflammation and PIH were assessed by facial examination, photographs, and questionnaires. Besides, PIH was evaluated by the Hyperpigmentation Activity and Severity Index (HASI) and a five-point grading system. RESULTS Moderate-to-severe PIH was observed in both groups. However, the mean HASI score and severity of PIH in the experimental group were lower than those of the control group at weeks 8 and 12 (P

Aim: This study examined the role of lymphocyte-to-monocyte ratio (LMR), an inflammatory biomarker, in predicting the severity of calcific aortic valve stenosis (CAVS) in a Chinese case–control study. Results: The LMR significantly decreased in the patients with CAVS compared with healthy controls. An inverse correlation was observed between the severity of stenosis and LMR in the patients. Additionally, the LMR was identified in the multivariate analysis as an independent predictor of severe CAVS. Conclusion: This study provides evidence of an inverse correlation between the severity of CAVS and LMR. LMR could potentially be applied as an independent predictor of severe CAVS and could be incorporated into a novel predictive model.

Information security has become an important subject in the artificial intelligence filed to handle big data. Most of the systems aim at obstacle detection on ordinary roads. In this paper, we prop...

The recommender systems have long been studied in the literature. The collaborative filtering is one of the most widely adopted recommendation techniques which is usually applied to the explicit data, e.g., rating scores. However, the implicit data, e.g., click data, is believed to be able to discover user’s latent preferences. Consequently, a number of research attempts have been made toward this issue. To the best of our knowledge, this paper is the first attempt to adapt the Wasserstein autoencoders to collaborative filtering problem. Particularly, we propose a new loss function by introducing an $$L_1$$ regularization term to learn a sparse low-rank representation form to represent latent variables. Then, we carefully design (1) a new cost function to minimize the data reconstruction error, and (2) the appropriate distance metrics for the calculation of KL divergence between the learned distribution of latent variables and the underlying true data distribution. Rigorous experiments are performed on three widely adopted datasets. Both the state-of-the-art approaches, e.g., Mult-VAE and Mult-DAE, and the baseline models are evaluated on these datasets. The promising experimental results demonstrate that the proposed approach is superior to the compared approaches with respect to evaluation criteria Recall@R and NDCG@R.

Early diagnosis of gastric cancer is difficult in China due to the lack of a valid method for endoscopic screening. Early gastric cancer, especially flat gastric cancer, lacks specific endoscopic features. Many cases appear to be similar to ordinary gastritis cases under normal white light endoscopy, which can lead to misdiagnosis. In order to find a new method to improve detection rate of early gastric cancer in China, we designed a trial to validate linked color imaging (LCI) for screening of early gastric cancer in a high-risk population, as compared to white light imaging (WLI). Subjects were randomly allocated to either the LCI + WLI or WLI group and then subjected to gastroscopy and all endoscopies were made after special preparation. All endoscopists had knowledge of this experiment. The main indicator was the rate of detection of gastric neoplastic lesions. The difference in the detection rate between the two groups is reported. The detection rate was 4.31% in the WLI group and 8.01% in the LCI + WLI group. This is a difference of 3.70% with a P value

Accurate and reliable fault diagnosis for rotating machinery, especially under variable working conditions remains a great challenge. Existing deep learning methods which extract features from single domain are insufficient to ensure reliable diagnosis results. In this study, a new deep learning based fault diagnosis method, which extracts features from both time and frequency domains is proposed. Two sets of deep features from multiple domains are fused into intrinsic low-dimensional features by local and global principle component analysis. And a new ensemble kernel extreme learning machine is proposed for fault pattern classification based on the fused features. Extensive experiments on gearbox, rotor and engine rolling bearing show that the proposed method has better diagnosis performance than state-of-the-art methods and is more adaptable to the fluctuation of working conditions.

Density peak clustering (DPC) is a recently developed density-based clustering algorithm that achieves competitive performance in a non-iterative manner. DPC is capable of effectively handling clusters with single density peak (single center), i.e., based on DPC’s hypothesis, one and only one data point is chosen as the center of any cluster. However, DPC may fail to identify clusters with multiple density peaks (multi-centers) and may not be able to identify natural clusters whose centers have relatively lower local density. To address these limitations, we propose a novel clustering algorithm based on a hierarchical approach, named multi-center density peak clustering (McDPC). Firstly, based on a widely adopted hypothesis that the potential cluster centers are relatively far away from each other. McDPC obtains centers of the initial micro-clusters (named representative data points) whose minimum distance to the other higher-density data points are relatively larger. Secondly, the representative data points are autonomously categorized into different density levels. Finally, McDPC deals with micro-clusters at each level and if necessary, merges the micro-clusters at a specific level into one cluster to identify multi-center clusters. To evaluate the effectiveness of our proposed McDPC algorithm, we conduct experiments on both synthetic and real-world datasets and benchmark the performance of McDPC against other state-of-the-art clustering algorithms. We also apply McDPC to perform image segmentation and facial recognition to further demonstrate its capability in dealing with real-world applications. The experimental results show that our method achieves promising performance. AI Singapore National Research Foundation (NRF) Accepted version This research is supported by the National Natural Science Foundation of China (61772227,61572227), the Science & Technology Development Foundation of Jilin Province (20180201045GX) and the Social Science Foundation of Education Department of Jilin Province (JJKH20181315SK). This research is also supported, in part, by the National Research Foundation Singapore under its AI Singapore Programme (Award Number: AISG- GC-2019-003), the Singapore Ministry of Health under its National Innovation Challenge on Active and Con dent Ageing (NIC Project No. MOH/NIC/COG04/2017), and the Joint NTU- WeBank Research Centre on Fintech, Nanyang Technological University, Singapore.

Modern approaches for semantic segmentation usually concatenate the feature map of the last convolutional layer and multi-scale features to form the final feature representation, which can achieve the more accurate classification of target pixels for the input image. However, the feature information of the last layer is not complete and refined so that there is a performance bottleneck in the concatenation between the final feature map and multi-scale feature representations. To solve this problem, we propose the Feature Recombining Network to get more refined and precise features for Semantic Segmentation. Our network is composed of Feature Recombining Module and Modified Pyramid Pooling Module. The two modules can extract more detailed and representative features through the feature recombination and acquire richer context information than the previous module respectively. Experiments show that our modules are effective to improve the segmentation precision and the Modified Pyramid Pooling Module is also superior to the previous module. Based on our proposed network, we achieve the performance of 51.9% mIoU on Pascal Context dataset and 44.75% mIoU on ADE20K dataset.

The thermal conductivity and thermal expansion coefficient are important thermo-physical properties for thermal barrier coatings (TBCs). In this work, multi-scale nanostructured 8YSZ coatings were fabricated using as-prepared nanostructured t′-8YSZ feedstocks by atmospheric plasma spraying (APS) onto NiCoCrAlYCe coated K417G superalloys. Meanwhile, conventional t-8YSZ feedstocks were also selected for comparison. The thermo-physical properties of both coatings have been studied. Results indicate that nanostructured 8YSZ coatings have a lower thermal conductivity than conventional counterparts especially when the temperature is above 1000 °C. Nanostructured 8YSZ coatings have better thermal insulation capability compared with conventional 8YSZ coatings. Therefore, the nanostructured coatings using t′-8YSZ feedstocks have a broad prospect for service at elevated temperature.

Iron supplementation is necessary for the treatment of anemia, one of the most frequent complications in inflammatory bowel disease (IBD). However, oral iron supplementation leads to an exacerbation of intestinal inflammation. Gut barrier plays a key role in the pathogenesis of IBD. The aim of this study was to characterize the interrelationship between systemic iron, intestinal barrier and the development of intestinal inflammation in a dextran sulfate sodium (DSS) induced experimental colitis mice model. We found that DSS-treated mice developed severe inflammation of colon, but became much healthy when intraperitoneal injection with iron. Iron supplementation alleviated colonic and systemic inflammation by lower histological scores, restorative morphology of colonic villi, and reduced expression of pro-inflammatory cytokines. Moreover, intraperitoneal supplementation of iron enhanced intestinal barrier function by upregulating the colonic expressions of tight junction proteins, restoring intestinal immune homeostasis by regulating immune cell infiltration and T lymphocyte subsets, and increasing mucous secretion of goblet cells in the colon. High-throughput sequencing of fecal 16 S rRNA showed that iron injection significantly increased the relative abundance of Bacteroidetes, which was suppressed in the gut microbiota of DSS-induced colitis mice. These results provided evidences supporting the protective effects of systemic iron repletion by intraperitoneal injection of iron on intestinal barrier functions. The finding highlights a novel approach for the treatment of IBD with iron injection therapy.

Plasmacytoid dendritic cells (pDCs) produce type I interferons (IFNs) and promote pathogenesis of multiple autoimmune diseases. Autoimmune Sjogren's syndrome (SS) primarily affects salivary and lacrimal gland, causing their inflammation, destruction and dysfunction. pDCs and type I IFN activity are elevated in salivary glands of SS patients, and this study seeks to elucidate the in vivo actions of pDCs in SS pathogenesis using the non-obese diabetic (NOD) mouse model. We confirmed the type I IFN-dependency of SS development in female NOD mice and elevation of pDC-type I IFN in their submandibular glands (SMGs). We administered a pDC-depleting anti-BST2/CD317 antibody to female NOD mice from 4 to 7 weeks of age, the early stage of SS, and assessed SS pathologies at age 10 weeks, the time of disease onset. Depletion of pDCs impeded the development of SMG inflammation and secretory dysfunction. It drastically reduced the amount of type I IFN mRNA and that of total leukocytes, and T- and B lymphocytes in SMGs. Gene expression analyses showed that pDC depletion markedly diminished SMG expression of IL-7, BAFF, TNF-α, IFN-γ, CXCL9, CXCL11, CD40, CD40L, Lt-α, Lt-β and NOS2. Hence, pDCs critically contribute to the development and onset of SS-like salivary gland exocrinopathy.

Background: Regular monitoring of serum potassium after total joint arthroplasty (TJA) is a routine examination, which can detect abnormal serum potassium and reduce adverse events timely caused by postoperative hypokalemia. In this study, we aimed to investigate the incidence and risk factors of hypokalemia after primary total hip and knee replacement.Methods: This study included patients who underwent unilateral total knee or hip arthroplasty in our department from April 2017 to March 2018. The serum potassium level before and after operation was collected and retrospectively analyzed. The differences in age, BMI and other factors between hypokalemia patients and non-hypokalemia patients at different time points after surgery were compared, and then the risk factors of postoperative hypokalemia were analyzed based on multiple logistic regression.Results: The total incidence of postoperative hypokalemia was 53.1%, while the respective rate on the first, third and fifth postoperative day was 12.5%, 40.7% and 9.6%. The serum potassium level on the first, third and fifth postoperative day was 3.84±0.32mmol/l, 3.59±0.34mmol/l and 3.80±0.32mmol/l, while among which, the level on the third day was the lowest (p=0.015). The independent risk factors for hypokalemia after total hip and knee replacement were the level of preoperative serum potassium (p=0.011), preoperative red blood cells (p=0.027), and a history of diabetes (p=0.007).Conclusion: Regular monitoring of serum potassium should be performed after TJA due to hypokalemia was a very common complication. We need to pay more attention to patients’ preoperative potassium and red blood cells, especially patients with diabetes.

Aim short-term blood pressure variability (BPV) as a risk factor of atherosclerosis and cardiovascular events has been investigated. However, its association with atherosclerotic plaque vulnerability remains unknown. The objective of this study was to determine the association between short-term BPV and intracranial atherosclerotic plaque vulnerability. Methods this is a cross-sectional analysis of 267 ischemic stroke patients with symptomatic intracranial atherosclerosis (mean age, 65±12 years old; 60.3% male), which were prospectively recruited in a comprehensive stroke center. Systolic and diastolic BP SD, CV, and BP variability ratio (BPVR) from 24 hours, daytime, and nighttime were calculated from 24-h ambulatory blood pressure monitoring, intracranial atherosclerotic plaque burden and vulnerability were evaluated by high-resolution magnetic resonance vessel wall imaging. Logistic regression analysis was used to locate the correlation between short-term BPV and plaque vulnerability. Results a total of 36.3% subjects presented with intraplaque hemorrhage (IPH) in this study. Multivariate logistic regression suggested that nighttime diastolic BP CV and 24-h BPVR were associated with intracranial IPH independently after adjusted for cardiovascular risk factors, odds ratio (OR) and 95% confidence interval (CI) for per SD BPV changes were 1.418 (1.051, 1.914) and 0.731 (0.548, 0.976), respectively, and this association also independent of atherosclerosis burden and 24-h mean systolic BP level. Further subgroup analysis by age and hypertension history demonstrated that the statistical correlation could only establish in the elder, and subjects with hypertension. Conclusion nighttime diastolic BP CV and 24-h BPVR were associated with intracranial IPH independently, especially in the elderly and subjects with hypertension.

Background: The aim of the present study was to investigate the influence of sagittal femoral bowing on sagittal femoral component alignment, and whether there was correlation between sagittal femoral component alignment and coronal femoral component alignment.Methods: We retrospectively reviewed 77 knees in 71 patients who had undergone primary TKA for advanced osteoarthritis.All surgeries were performed by using a standard medial parapatellar approach. The osteotomy was performed with a conventional technique using an intramedullar rod for the femur and a mechanical extramedullar guiding system for the tibia. All patients enrolled in the study were evaluated with full-length lower extremity load-bearing standing scanograms and the patients had preoperative and postoperative radiographs of the knees. cFBA(coronal femoral bowing angle), sFBA(sagittal femoral bowing angle),and postoperatively, mTFA(mechanical tibiofemoral angle of the knee), β angle(femoral component flexion angle) were measured. The radiographic results of both groups were compared using Student's t test. A two-sided Pearson correlation coefficient was obtained to identify the correlations between FBA in the coronal and sagittal planes, as well as FBA and age or BMI, sFBA and β angle, cFBA and mTFA. Comparison of FSB incidence between different genders was using chi-square test. The p value Results: The mean sFBA, cFBA, β angle, mTFA were 9.34°±3.56°(range 1°-16°), 3.25°±3.79°(range -7°-17°), 3.91°±3.15°(range -1°-13°), 0.60°±1.95°(range -3°-6°), respectively. There was no correlation between age and sFBA(CC=0.192, p=0.194) or cFBA(CC=0.192, p=0.194), similarly, there was no correlation between age and sFBA(CC=0.067, p=0.565) or cFBA(CC=0.069, p=0.549). The sFBA was correlated with cFBA and β angle(CC=0.540, pppp=0.247) of cFBA between the patients with sFSB and the patients without sFSB. Conclusions: The current study showed that the sFBA was correlated with cFBA in the patients undergoing TKA and the patients with sFSB usually presented non-cFSB. We also found that sFSB could affect the femoral component alignment in the sagittal plane and cFSB could affect the femoral component alignment in the coronal plane. The sFBA or cFBA was not correlated with age, BMI or gender.

Vortex field can be generated by the phase modulated circular transducer array. However, the side lobe of the beam may interfere the transmission and reception of the signals. In order to reduce the influence of side lobe, a phased modulated concentric ring array is proposed in this paper to produce the vortex field with low side lobe. By exploring the far-field radiation patterns of the concentric ring array with different inner circle radius, the number of array elements and the initial phase, we obtain the suitable array arrangement with better radiation pattern. This paper should be of positive significance to the development of underwater sound field detection and identification technology.

The purpose of this study was to evaluate the clinicopathological features of different degrees of extraglomerular renal vascular lesions (RVLs) in patients with anti-neutrophil cytoplasmic antibody (ANCA)-associated renal vasculitis and explore their clinical determinants. This is a retrospective study of 186 patients with ANCA-associated renal vasculitis diagnosed at the First Affiliated Hospital of Zhengzhou University from January 2014 to April 2019. The patients who met the inclusion criteria were divided into non-renal RVLs, mild RVLs, moderate RVLs, and severe RVLs. It was found that there were significant differences in serum creatinine (SCR), estimated glomerular filtration rate (eGFR), erythrocyte sedimentation rate (ESR), high-density lipoprotein (HDL), systolic blood pressure (SBP), the prevalence rate of hypertension, the proportion of normal glomeruli, and the proportion of sclerotic glomeruli and interstitial fibrosis integral. SCR and ESR are independent risk factors for RVLs. The participants were followed up for 1 year, and the progression to end-stage renal disease (ESRD) and death was defined as endpoint events. We found that the survival rate of patients without RVLs was significantly higher than that of patients with RVLs and that the RVLs were an independent risk factor for ESRD or death. Early intervention in the progression of RVLs can improve the prognosis.

High on-treatment platelet reactivity (HOPR) is associated with increased risk of cardiovascular events in patients undergoing percutaneous coronary intervention (PCI). We randomised post-PCI patients with HOPR after 5 days of standard dual antiplatelet therapy (DAPT) to intensified therapy with aspirin 100 mg once daily in combination with either clopidogrel 150 mg once daily, clopidogrel 75 mg once daily plus cilostazol 100 mg twice daily, ticagrelor 90 mg twice daily, or standard therapy with clopidogrel 75 mg once daily (STD) for 1 month, after which all patients were switched to standard DAPT for a further 11 months. The primary outcome was residual HOPR rate at 1 month. We screened 1724 patients with light transmission aggregation studies and randomised 434 with HOPR. At 1 month the proportion of patients with persistent HOPR was significantly lower in the intensified therapy groups compared with STD group. Compared to the group receiving STD therapy, those receiving intensified therapy had significantly lower rate of major adverse cardiovascular events (MACE) at both 1 month and 12 months with no significant increase in bleeding. In patients with post-PCI HOPR, 1 month of intensified antiplatelet therapy provides greater platelet inhibition and improves outcomes without increasing bleeding. Clinical Trial Registration URL: http://www.clinicaltrials.gov; Unique Identifier: NCT01955200.

Backgrounds Pregnancy termination during the second trimester in patients with placenta previa and placenta accreta spectrum (PAS) is a complex and challenging clinical problem. Based on our literature review, there has been a relative increase in the number of such cases being treated by hysterotomy and/or local uterine lesion resection and repair. In the present study, a retrospective analysis was conducted to compare the clinical outcomes when different management strategies were used to terminate pregnancy in the patients with placenta previa and PAS. Methods A total of 51 patients who underwent pregnancy termination in the second trimester in Beijing Obstetrics and Gynecology Hospital between June 2013 and December 2018 were retrospectively analyzed in this study. All patients having previous caesarean delivery (CD) were diagnosed with placenta previa status and PAS. Results ① Among the 51 patients, 16 cases received mifepristone and misoprostol medical termination, 15 cases received mifepristone and Rivanol medical termination, but 1 of them was transferred to hysterotomy due to failed labor induction, another 20 cases were performed planned hysterotomy. There was no placenta percreta cases and uterine artery embolization (UAE) was all performed before surgery.② There were 31 cases who underwent medical termination and 30 cases were vaginal delivery. Dilation and evacuation (D&E) were used in 20 cases of medical abortion failure and in all 30 cases of difficult manual removal of placental tissue. ③ A statistically significant difference was found among the three different strategies in terms of gestational weeks, the type of placenta previa status, main operative success rate and β-HCG regression time (P Conclusions Terminating a pregnancy by vaginal delivery through medical induction of labor may be feasible if clinicians have an overall understanding of gestational age, the type of placenta previa status, the type of placenta accreta, and patients concerns about preserving fertility. A collaborative team effort in tertiary medical centers with a very experience MDT and combined application of multiple methods is required to optimize patient outcomes.

The aerial part of Biebersteinia heterostemon Maxim. (Geraniaceae Biebersteiniaceae) known as ming jian na bao in Chinese, has been traditionally used in Tibetan folk medicine for treatment of diabetes and hypertension. The aim of the present study was to evaluate the effects of galegine obtained from an ethanol extract of the entire Biebersteinia heterostemon plant on the rat’s cardiovascular system in order to characterize its contributions as an antihypertensive agent. The antihypertensive effect of galegine was investigated in pentobarbital-anesthetized hypertensive rats at three dose levels based on the LD50 of galegine. Meanwhile a positive control group received dimaprit with the same procedure. Dimaprit infusion induced a significant hypotension which declined by an average margin of 20%. Simultaneously, single administration of galegine at the doses of 2.5, 5, and 10 mg/kg by intraperitoneal injection induced an immediate and dose-dependent decrease in mean arterial blood pressure (MABP) by an average margin of 40% with a rapid increase in heart rate (HR). We demonstrated that galegine is effective in reducing blood pressure in anesthetized hypertensive rats with rapid onset and a dose-related duration of the effects. The results indicate that galegine was the bioactive compound which can be used as a pharmacophore to design new hypertensive agents.

Background Robotic-assisted total knee arthroplasty (TKA) was performed to promote the accuracy of bone resection and mechanical alignment. Among these TKA system procedures, 3D reconstruction of CT data of lower limbs consumes significant manpower. Artificial intelligence (AI) algorithms applying deep learning has been proved efficient in automated identification and visual processing. Methods CT data of a total of 200 lower limbs scanning were used for AI-based 3D model construction and CT data of 20 lower limbs scanning were utilised for verification. Results We showed that the performance of an AI-guided 3D reconstruction of CT data of lower limbs for robotic-assisted TKA was similar to that of the operator-based approach. The time of 3D lower limb model construction using AI was 4.7 min. AI-based 3D models can be used for surgical planning. Conclusion AI was used for the first time to guide the 3D reconstruction of CT data of lower limbs for facilitating robotic-assisted TKA. Incorporation of AI in 3D model reconstruction before TKA might reduce the workload of radiologists.

With the development of convolutional neural networks (CNN), the super-resolution results of CNN-based method have far surpassed traditional method. In particular, the CNN-based single image super-resolution method has achieved excellent results. Video sequences contain more abundant information compare with image, but there are few video super-resolution methods that can be applied to mobile devices due to the requirement of heavy computation, which limits the application of video super-resolution. In this work, we propose the Efficient Video Super-Resolution Network (EVSRNet) with neural architecture search for real-time video super-resolution. Extensive experiments show that our method achieves a good balance between quality and efficiency. Finally, we achieve a competitive result of 7.36 where the PSNR is 27.85 dB and the inference time is 11.3 ms/f on the target snapdragon 865 SoC, resulting in a 2nd place in the Mobile AI (MAI) 2021 real-time video super-resolution challenge. It is noteworthy that, our method is the fastest and significantly outperforms other competitors by large margins.

During the process of China's rural urbanization and agricultural supply-side reform, the disposal of a large number of agricultural and forestry wastes will be a big problem. This study deals with effect of four variables on the thermal, mechanical, hydraulic properties and microscopic morphology of the bio-insulation materials using wheat straw as aggregate and geopolymer as binder. The four variables are wheat straw length dimension (L), prewetting water /wheat straw mass ratio1 (Wp/W), wheat straw/solid content of binder mass ratio (W/S) and H2O2/solid content of binder mass ratio (H/S), respectively. The results show that all the four variables have a great influence on the foaming process and skeleton structure of the geopolymer, which greatly affect the properties of the bio-insulation materials. In conclusion, the thermal conductivity of bio-insulation materials varies from 0.092 to 0.186 W m−1 K−1, and the compressive strength varies from 0.18 to 5.622 MPa with the density being 235–894.1 kg m−3. Moreover, the values of water and moisture absorption rates are 32–107% and 4–33% respectively. A desirable bio-insulation material with wheat straw length of 0.2 cm, Wp/W being 1, W/S being 0.124, and H/S being 0.0124, is recommended, of which the thermal conductivity is 0.101 W m−1 K−1 with the compressive strength being 1.2 MPa and the density being 302.4 kg m−3. This new kind of bio-insulation material exhibits good thermal and mechanical performances, which allows it to be applied as a thermal insulation material in buildings, especially the prefabricated buildings. And it is beneficial to energy saving and environmental protection.

Image segmentation is the basis of image analysis, object tracking, and other fields. However, image segmentation is still a bottleneck due to the complexity of images. In recent years, fuzzy clustering is one of the most important selections for image segmentation, which can retain information as much as possible. However, fuzzy clustering algorithms are sensitive to image artifacts. In this study, an improved image segmentation algorithm based on patch-weighted distance and fuzzy clustering is proposed, which can be divided into two steps. First, the pixel correlation between adjacent pixels is retrieved based on patch-weighted distance, and then the pixel correlation is used to replace the influence of neighboring information in fuzzy algorithms, thereby enhancing the robustness. Experiments on simulated, natural and medical images illustrate that the proposed schema outperforms other fuzzy clustering algorithms.

BACKGROUND Melasma is a benign and chronic hypermelanosis characterized by irregular light brown to dark brown patches of hyperpigmentation on the skin. Oral tranexamic acid (TA) or vitamin C (VC) supplementation has been one treatment choice. TA interferes with keratinocyte-melanocyte interactions, and VC functions by reducing melanin production resulting in skin rejuvenation and whitening. AIM The aim of the present study was to compare the efficacy and safety of Myjet assisted transdermal injection of TA vs VC in the treatment of melasma. METHODS In this split-face controlled trial, 17 patients were randomized to receive eight weekly transdermal injections of TA or VC via Myjet either on the right or the left side of their face. MASI was measured from each side of the face at the baseline, at the middle, and at the end of treatment. RESULTS A reduction in MASI was observed for TA and VC separately (P value

In this research, energy level based exergoeconomic evaluations are performed for a novel biomass and natural gas fired polygeneration system of electricity, hot water, chilled water and hydrogen production. The proposed system mainly consists of a biomass gasifier, a proton exchange membrane (PEM) electrolyzer, a gas turbine cycle (GT), an absorption chiller, and a ground source heat pump cycle. Conventional and energy level based exergoeconomic performances of the proposed system are compared; related exergy and economic analysis are also performed. In addition, the variations in unit exergy cost of products (electricity, hot water, chilled water and hydrogen) are studied under economic factors. The results show that energy and exergy efficiency of the electrolyzer and the proposed system decrease with the increasing current density of the PEM electrolyzer. The unit exergy cost of electricity and hydrogen are 5.24 $/GJ and 20.41 $/GJ under the energy level based exergoeconomic method, respectively, which are higher than that under the conventional exergoeconomic method (electricity: 4.38 $/GJ, hydrogen: 19.00 $/GJ), while the unit exergy cost of hot water and chilled water under the energy level based exergoeconomic method are lower than that under the conventional exergoeconomic method. Moreover, the exergoeconomic factor and relative cost difference of the system equipment also show distinctions under the conventional and energy level based exergoeconomic methods. The presented polygeneration system is a promising technology to utilize renewable energy and improve the flexibility of the integrated system; and the energy level based exergoeconomic method shows certain rationality and feasibility in the system analysis.

Road information extraction based on aerial images is a critical task for many applications, and it has attracted considerable attention from researchers in the field of remote sensing. The problem is mainly composed of two subtasks, namely, road detection and centerline extraction. Most of the previous studies rely on multistage-based learning methods to solve the problem. However, these approaches may suffer from the well-known problem of propagation errors. In this paper, we propose a novel deep learning model, recurrent convolution neural network U-Net (RCNN-UNet), to tackle the aforementioned problem. Our proposed RCNN-UNet has three distinct advantages. First, the end-to-end deep learning scheme eliminates the propagation errors. Second, a carefully designed RCNN unit is leveraged to build our deep learning architecture, which can better exploit the spatial context and the rich low-level visual features. Thereby, it alleviates the detection problems caused by noises, occlusions, and complex backgrounds of roads. Third, as the tasks of road detection and centerline extraction are strongly correlated, a multitask learning scheme is designed so that two predictors can be simultaneously trained to improve both effectiveness and efficiency. Extensive experiments were carried out based on two publicly available benchmark data sets, and nine state-of-the-art baselines were used in a comparative evaluation. Our experimental results demonstrate the superiority of the proposed RCNN-UNet model for both the road detection and the centerline extraction tasks.