Search

There is a growing demand for multi-port power converters for managing multiple energy sources and loads. To that end, a dual-input single-output dc-dc converter is proposed in this article. The converter integrates two different and low-voltage dc sources into a dc bus. Pulse width modulation is used to control the transfer energy from the input ports to the output port. The discontinuous conduction mode (DCM) operation of the converter has enabled the soft-switching operation in the primary switches. Moreover, it helps avoid the high-frequency transformer (HFT) saturation when the two input voltage sources have different values. The detailed analysis of the proposed dc-dc converter is presented in this article along with the simulation and experimental results obtained from a 150 W hardware prototype to validate its operation.

AC-type flying capacitor concept has been proposed for current source converter (CSC) topologies recently, which offers unique merits over traditional CSCs, such as lower dv/dt with improved DC-link voltage quality as well as flexible common-mode voltage (CMV) control. To cope with main-stream three-phase industrial applications, the three-phase flying-capacitor-clamped CSC (FCC-CSC) is introduced and extensively investigated in this paper for the first time. A novel space vector modulation (SVM) is proposed, featuring a modular design process, active flying capacitor voltage control, and CMV suppression. The detailed commutation principles and the capacitor clamping voltage influence under different switching states are comprehensively addressed. Then, the redundant switching states are optimized to fulfill the multiple targets simultaneously. Simulation and experimental results are performed to demonstrate the superior dc-link power quality and CMV performance compared to traditional three-phase CSC topology.

Smart space has emerged as a new paradigm that combines sensing, communication, and artificial intelligence technologies to offer various customised services. A fundamental requirement of these services is person identification. Although a variety of person identification approaches have been proposed, they suffer from several limitations in practical applications such as low energy efficiency, accuracy degradation, and privacy issue. This paper proposes an energy harvesting based privacy-preserving gait recognition scheme for smart space which is named . In , we extract discriminative features from one-dimensional gait signal and design an attention-based long short term memory (LSTM) network to classify different people. Moreover, we leverage a novel Bloom filter-based privacy-preserving technique to address the privacy leakage problem. To demonstrate the feasibility of , we design a proof-of-concept prototype using off-the-shelf energy harvesting hardware. Extensive evaluation results show that the proposed scheme outperforms state-of-the-art by 6–10% and incurs low system cost while preserving user’s privacy.

Ring and meshed dc distribution grids have attracted increasing interest lately. Their multi-point and multi-terminal structure facilitates flexible operational control and reliable power supply across the network. However, the systems implemented with existing power converters lack sufficient control freedoms for proper power flow adjustment; therefore, it is necessary to develop additional dc power flow controllers (DCPFCs) to be added to these grids for improved power flow control capability. Specific to multi-terminal dc networks built with modular multilevel converters (MMCs), this paper proposes an embedded DCPFC (e-DCPFC) directly integrated in the MMC topology. Unlike previously proposed solutions, the e-DCPFC configuration is derived from the conventional MMC and does not need any isolation transformers or external power supply. Its other main features include modular construction, convenience for multiport extension and bipolar regulating capacity. In this paper, the topology, working principle, and the coordinated control strategy for e-DCPFC and its host MMC are elaborated. Experimental results from a lab prototype verify the proposed topology and its performance under typical working conditions.

Lung cancer is the top reason for cancer-related deaths worldwide. The 5-year overall survival rate of lung cancer is approximately 20 % due to the delayed diagnosis and low response rate to regular treatments. Microbiota, both host-microbiota and alien pathogenic microbiota, have been investigated to be involved in a complicated and contradictory relationship with lung cancer initiation, treatments, and prognosis. Disorders of certain host-microbiota and pathogen infection are associated with the risk of lung cancers based on epidemiological evidence, and antibiotics (ATBs) could dramatically impair anti-cancer treatment efficacy, including chemotherapy and immunotherapy. Moreover, probiotics and microbe-mediated drugs are potential approaches to enhance regular anti-tumor treatments. Therefore, the knowledge of the complex dual effect of microbes on lung cancer is beneficial to take their essence and remove their dross. This review offers insight into the current trends and advancements in microbiota or microbial components related to lung cancer.

The issue of Electroencephalogram (EEG)-based emotion recognition has great academic and practical significance. Currently, there are numerous researches trying to address this issue in the literature. Particularly, transfer learning has gradually become a new methodological trend for the issue in company with the popularity of deep learning. Motivated by capturing the research panorama, summarizing the technological essence, and forecasting the advancement tendency of transfer learning for EEG-based emotion recognition, this paper contributes a review work. This work mainly includes five aspects: a) introducing the issue of EEG-based emotion recognition and expounding the importance of transfer learning for it; b) analyzing the transfer learning framework and comparing it with the traditional ones; c) elucidating the issue difficulties and explaining the suitability and capability of transfer learning for this issue; d) summarizing, categorizing and exemplifying the typical transfer learning methods for this issue; e) clarifying the methodological merits, discussing the challenging problems, and predicting the prospective development of transfer learning for the issue. We expect these contributions can inspire innovation and reformation of the transfer learning methodology for EEG-based emotion recognition as well as other relevant topics in the not-so-far future.

Dynamic ridesharing refers to services that arrange one-time shared rides on short notice. It underpins various real-world intelligent transportation applications such as car-pooling, food delivery and last-mile logistics. A core operation in dynamic ridesharing is the "insertion operator". Given a worker and a feasible route which contains a sequence of origin-destination pairs from previous requests, the insertion operator inserts a new origin-destination pair from a newly arrived request into the current route such that certain objective is optimized. Common optimization objectives include minimizing the \rev{maximum/sum} flow time of all requests and minimizing the total travel time of the worker. Despite its frequent usage, the insertion operator has a time complexity of $O(n^3)$ , where $n$ is the number of all requests assigned to the worker. The cubic running time of insertion fundamentally limits the efficiency of urban-scale dynamic ridesharing based applications. In this paper, we propose a novel partition framework and a dynamic programming based insertion with a time complexity of $O(n^2)$ . We further improve the time efficiency of the insertion operator to $O(n)$ harnessing efficient index structures, such as fenwick tree. Evaluations on two real-world large-scale datasets show that our methods can accelerate insertion by 1.5 to 998.1 times.

Abstract: miRNA-129-5p belongs to the microRNA-129 (miRNA-129) family. miRNA-129-5p is expressed in many tissues and organs of the human body, and it regulates a wide range of biological functions. The abnormal expression of miRNA-129-5p is related to the occurrence and development of a variety of malignant tumors. miRNA-129-5p plays an important role in the tumorigenesis process and functions by promoting or inhibiting tumors. However, the role of miRNA-129-5p in cancer remains controversial. This article reviews the different biological functions of miRNA- 129-5p in cancer and provides ideas for research in this field to guide the development of targeted therapies and drugs for malignant tumors.

As the ocean engineering structure serves in the critical underwater environment, a variety of defects, such as cracks and corrosions, always cause damage to the structure. It is still a key challenge to identify and evaluate these defects accurately in the underwater environment. In this paper, the visual and intelligent identification methods are presented for the inspection of defects in underwater structures using the alternating current field measurement (ACFM) technique. The current perturbation theory is developed to analyze the disturbed current field and the distorted magnetic field caused by defects. The gradient imaging algorithm is presented as an image preprocessing method to highlight the visual morphology of defects. The underwater intelligent ACFM system is set up. The experiments are carried out to verify the gradient imaging algorithm. The convolutional neural network (CNN) deep learning algorithm is presented to identify the grey-scale map samples preprocessed by the gradient imaging algorithm. The results show that the current perturbation theory clarifies the relationship between the characteristic signal and the morphology of various defects. The Bz image reflects the surface morphology of defects. The gradient imaging algorithm can achieve visual detection of defects. The single crack, the irregular crack and the corrosion can be identified intelligently by the CNN deep learning algorithm. These defects can be evaluated accurately after classification.

Due to the unpredictable and fluctuating nature of solar photovoltaic (PV), energy storage systems (ESS), such as batteries, are always integrated with PV systems to smooth the power supply. In this paper, a multi-port dc-ac converter (MPC) with differential power processing dc-dc converter (DPPC) is proposed for battery ESS integrated PV systems. The MPC is capable of regulating most of active power among PV, battery and ac grid, and only the differential power (partial power) needs to be handled by the dc-dc converter. Hence, the major merits of higher integration, higher efficiency, and lower cost can be achieved by the proposed configuration. A new cooperative control scheme for the MPC and DPPC is investigated, aiming at realizing flexible active power flow. Besides, a modified space vector pulse-width modulation (SVPWM) is developed for the MPC, taking into the consideration of the voltage variation of both PV and battery. The active power controllability of the MPC and the power rating of the DPPC are quantitatively analyzed. The effectiveness of the proposed configuration, control and modulation schemes is validated by experimental results.

Gastric cancer is a global cancer with a high mortality rate. A growing number of studies have found the abnormal expression of lncRNA (long noncoding RNA) in many tumors, which plays a role in promoting or inhibiting cancer. Similarly, lncRNA abnormal expression plays an essential biological function in gastric cancer. This article focuses on lncRNA involvement in the development of gastric cancer in terms of cell cycle disorder, apoptosis inhibition, metabolic remodeling, promotion of tumor inflammation, immune escape, induction of angiogenesis, and Epithelial Mesenchymal Transition (EMT). The involvement of lncRNA in the development of gastric cancer is related to drug resistance, such as cisplatin and multi-drug resistance. It can also be used as a potential marker for the diagnosis and prognosis of gastric cancer and a target for the treatment. With an in-depth understanding of the mechanism of lncRNA in gastric cancer, new ideas for personalized treatment of gastric cancer are expected.

BACKGROUND: PPM1G, a member of the serine/threonine protease family, dephosphorylates various proteins and may be involved in cancer development. The role and mechanism of PPM1G in HCC still needs to be verified. OBJECTIVE: This study aims to explore the role of PPM1G in the occurrence, development and prognosis of HCC. METHODS: Using bioinformatics (UALCAN, cBioPortal, Linkedomics, STRING and GSEA) to analyze the expression of PPM1G mRNA in HCC, its clinical relevance and possible involved signaling pathways. The expression of PPM1G protein was determined by immunohistochemistry in 311 cases of HCC to evaluate the association between PPM1G and clinical features and prognosis. RESULTS: The expression of PPM1G was significantly upregulated in HCC (P< 0.001), correlated with the metastasis (P= 0.020), pathological grade of HCC (P= 0.032), microvascular invasion (P= 0.040), and HBV infection (P= 0.041). Cox multivariate regression showed high expression of PPM1G was an independent prognostic factor for HCC. Its role in HCC may relate to methylation and frequency mutation. Furthermore, the database showed PPM1G is involved in the signal pathway such as cell cycle, WNT pathway, and mTOR pathway in HCC. CONCLUSION: PPM1G showed an essential function involving in tumor-related pathways in HCC, providing a biological basis for targeted treatment of HCC clinically.

With the rapid development of microwave photonics technology, high-speed processing and ultra-weak signal detection capability have become the main bottlenecks in many applications. Thanks to the ultra-weak signal detection capability and the extremely low timing jitter properties of single-photon detectors, the combination of single-photon detection and classical microwave photonics technology may provide a solution to break the above bottlenecks. In this paper, we first report a novel concept of single-photon microwave photonics (SP-MWP), a SP-MWP signal processing system with phase shifting and frequency filtering functionalities is demonstrated based on a superconducting nanowire single photon detector (SNSPD) and a successive time-correlated single photon counting (TCSPC) module. Experimental results show that an ultrahigh optical sensitivity down to −100 dBm has been achieved, and the signal processing bandwidth is only limited by the timing jitter of single-photon detectors. In the meantime, the proposed system demonstrates an ultrahigh anti-interference capability, only the signal which is phase locked by the trigger signal in TCSPC can be extracted from the detected signals combining with noise and strong interference. The proposed SP-MWP concept paves a way to a novel interdisciplinary field of microwave photonics and quantum mechanism, named by quantum microwave photonics.

OBJECTIVES To compare drug-coated balloon (DCB) and bare metal stent (BMS) for primary lesions in femoropopliteal artery disease in Chinese population and to make subgroup analysis between the groups. METHODS Patients with primary lesions who underwent BMS or DCB treatment of a single tertiary vascular center were included and followed up for 24 months. Clinical and anatomic status were reported using the criteria recommended by the Society for Vascular Surgery. The primary endpoint included primary patency, clinically target limb revascularization, composite safety endpoint and all-cause death over 24 months assessed by Kaplan-Meier. Secondary endpoints included technical success rate and stent-related complications. RESULTS 284 patients with 324 limbs were pooled into analysis and most of the baseline characteristics did not show significant difference. 74 in BMS group and 71 in DCB group were claudicants while 83 in BMS group and 56 in DCB group suffered from chronic limb threatening ischemia (CLTI). The mean cumulative lesion length was 18.7±9.8cm in BMS group while 17.2±10.3cm in DCB group. Kaplan-Meier estimates of primary patency were 75.3% and 80.9% for BMS and DCB groups at 12 months while decreased to 63.9% and 70.2% at 24 months (log-rank p=0.167), respectively. Freedom from clinically driven target limb revascularization was 86.8% and 92.7% for BMS and DCB groups at 12 months while dropped to 82.5% and 85.9% at 24 months (log-rank p=0.342). Estimates of primary patency between BMS and DCB group did not show significant difference on lesions with poor runoff (58.8% versus 67.3%, log-rank p=0.127), severe calcification (64.5% versus 69.4%, log-rank p=0.525) and popliteal artery involvement (59.3% versus 60.3%, log-rank p=0.695) at 24 months. The overall survival (92.6% for BMS, 90.3% for DCB, log-rank p=0.391) and freedom from composite safety endpoint (79.3% for BMS, 79.2% for DCB, log-rank p=0.941) showed no significant difference at 24 months. CONCLUSIONS Over the 24-month follow-up, BMS and DCB showed equivalent efficacy and safety outcomes for primary femoropopliteal artery disease, which indicated the reduction of permanent metallic implant insertion might be possible.

Adverse weather conditions seriously threaten the traffic safety, especially for rainy days with the ponding water on the road surface, which potentially result in vehicle crashes, person injuries and crash fatalities. Automatic splashed water detection based on surveillance videos is an attractive way to effectively prevent the traffic accidents. However, surveillance videos exhibit great variations with lighting changes, illumination conditions and complex backgrounds, which pose great difficulties in automatic recognition. In this paper, a novel deep learning based approach is proposed to detect the splashed water. To the best of our knowledge, this is the first work on this topic based on deep learning. An effective semantic segmentation network, called SWNet, is novelly proposed to extract the potential splashed water regions. An encoder-decoder structure is designed to capture the visual characteristics of splashed water. SWNet achieves high efficiency by reusing pooling indices and adopting the light-weight decoder. With the multi-scale feature fusion structure, SWNet integrates the coarse semantic information and detailed appearance information, which significantly boosts the accuracy and refines the edge segmentation. A weighted cross entropy loss for splashed water is adopted to cope with the unbalanced distribution between splashed water and backgrounds. Moreover, a splashed water attention module is designed to focus on the salient regions of moving vehicles and splashed water, by performing attention mechanism to integrate global contextual information in semantic segmentation. Experiments conducted on a newly collected splashed water dataset demonstrate the effectiveness and efficiency of the proposed approach, which outperforms the state-of-the-art methods.

Limited by the computational capabilities and battery energy of terminal devices and network bandwidth, emotion recognition tasks fail to achieve good interactive experience for users. The intolerable latency for users also seriously restricts the popularization of emotion recognition applications in the edge environments such as fatigue detection in auto-driving. The development of edge computing provides a more sustainable solution for this problem. Based on edge computing, this article proposes a multi-modal multi-layer emotion-aware service (MULTI-EASE) architecture that considers user's facial expression and voice as a multi-modal data source of emotion recognition, and employs the intelligent terminal, edge server and cloud as multi-layer execution environment. By analyzing the average delay of each task and the average energy consumption at the mobile device, we formulate a delay-constrained energy minimization problem and perform a task scheduling policy between multiple layers to reduce the end-to-end delay and energy consumption by using an edge-based approach, further to improve the users' emotion interactive experience and achieve energy saving in edge computing. Finally, a prototype system is also implemented to validate the architecture of MULTI-EASE, the experimental results show that MULTI-EASE is a sustainable and efficient platform for emotion analysis applications, and also provide a valuable reference for dynamic task scheduling under MULTI-EASE architecture.

Heat-integrated water network synthesis (HIWNS) has received considerable attention for the advantages of reducing water and energy consumptions. HIWNS is effective in water and energy sustainability. Mixed integer non-linear programming (MINLP) is usually applied in HIWNS. In this work, a novel nonlinear programming (NLP) was proposed for HIWNS by considering wastewater reuse and wastewater regeneration reuse. Integer variables are changed to non-linear equation by the methods for identifying stream roles and denoting the existence of process matches. The model is tested by examples with single and multiple regeneration unit problems. The testing results showed that the NLP is an alternative method for HIWNS with wastewater reuse and regeneration reuse.

A dual-band tunable balanced filter based on one hybrid microstrip-cavity structure is proposed in this paper. The cavity acts as one multimode resonator, and the differential feeding networks are printed on the PCBs. According to the electromagnetic (EM) field theory, TE101 and TE011 modes are generated under differential-mode (DM) excitation, and highly suppressed under common-mode (CM) excitation. The dielectric posts in the cavity can independently shift the resonant frequencies of the two modes. Moreover, the inter-resonator coupling of each band is independently controlled by the cross-shaped coupling slot. Therefore, good filtering differential response can be realized with in a wide band. For demonstration, a dual-band tunable balanced filter is designed and measured. The measured results indicate that the center frequencies of the two bands are tuned from 7.51 GHz to 8.22 GHz and 9.34 GHz to 10.56 GHz, respectively. Over the tuning range, the insertion losses vary from 0.5 dB to 1.5 dB and the CM suppressions in both passbands are under 50 dB and 42 dB. In this way, the proposed hybrid microstrip-cavity structure is feasible to design a reconfigurable balanced filter with wide tuning range and high CM suppression.

Location-based services have become more and more important with the development of the Internet of Things (IoT) technology in recent years. Global Navigation Satellite System (GNSS) is widely used for positioning outdoors while it is still challenging to realize accurate and universal 3D indoor localization in complex indoor environments. Crowdsourcing-based positioning method is proposed aiming at autonomously constructing the navigation database based on the pedestrians’ daily-life data. This paper proposes an autonomous 3D indoor localization algorithm using the combination of crowdsourced Wi-Fi fingerprinting and Micro-Electro-Mechanical System sensors (3D-CSWS). An enhanced complementary filter is applied to provide accurate attitude information by integrating multi-sensor data with the detection of external acceleration and quasi-static magnetic field. In addition, the gradient descent (GD) algorithm is proposed to optimize the forward pedestrian dead reckoning and the optimized trajectories are weighted fused to construct the final navigation database after quality evaluation. In the on-line phase, the adaptive particle filter is used to integrate the results of Wi-Fi fingerprinting and multiple sensors to provide accurate and concrete 3D indoor localization performance. The real-world experimental results demonstrate that the proposed 3D-CSWS is proved to achieve autonomous and precise 3D indoor localization performance among complex indoor environments.

Background Hand foot and mouth disease (HFMD) is a public health concern. Studies on air pollution and the lengths of hospital stay (LOS) of HFMD are scarce. Methods We characterized the clinic demographic features of 5135 hospitalized HFMD in Nanjing, China from 2012 to 2017, and analyzed the association between short-term exposure to PM2.5 as well as its components (OM, BC, SO42−, NH4+, NIT, SOIL and SS) and the LOS of HFMD. Results Among them 98.62% were aged 0–6 years old, and 3772 (73.46%) were hospitalized for seven days or less. The LOS of HFMD children was different in different ages, illness onset years and illness onset seasons (P2.5 concentrations, LOS of HFMD increased by 0.52(0.33, 0.71), 0.50 (95% CI, 0.31–0.69) and 0.46 (95% CI, 0.28–0.65) day in adjusted models at lag 3 days, lag 7days and lag 14 days, respectively. In addition, per IQR increase of BC, SO42−, NH4+, NIT and SOIL were also significantly associated with the LOS of HFMD. Conclusions Our findings corroborated the hypothesis that short-term PM2.5 exposure was associated with increased the LOS of HFMD, and its components (BC, SO42−, NH4+, NIT and SOIL) of PM2.5 might play a key role in the prolonged LOS of HFMD.

Objective Previous clinical studies and meta-analyses have shown controversial results on the association between C3435T polymorphism of the ABCB1 gene and anti-epileptic drug (AED) resistance. Based on the fact that sample size and confounding factors could contribute to the inconsistency, we performed an updated meta-analysis by including the most recent studies, and subgroup analysis was conducted to evaluate the effect of confounding factors on the association. Materials and methods We searched articles in 6 electronic databases including PubMed, Medline, Embase, Web of science, Cochrane Library, CNKI (China National Knowledge Infrastructure) for relevant articles up to June 2020. Results The current analysis showed that the C allele of C3435T variant was a risk factor for drug resistance in the overall populations (C allele vs. T allele, OR: 1.13; 95% CI: 1.02 - 1.25; p = 0.02) and in the Caucasians (C allele vs. T allele, OR: 1.09; 95% CI: 1.09 - 1.43; p = 0.002), while no association was observed in Asians and Indians. Particularly, our study reported for the first time that the 3435T allele was more common in epilepsy patients with drug resistance in the Tunisian population (C allele vs. T allele, OR: 0.31; 95% CI: 0.15 - 0.65; p = 0.002). In addition, our present analysis suggested an association between C3435T and AED resistance in cryptogenic, symptomatic, but not in idiopathic patients. Subgroup studies based on age and gender showed no association. Conclusion AED resistance in Caucasian and Tunisian populations may benefit from ABCB1 C3435T genotyping. We recommend that more details, such as gender and etiology of epilepsy, should be taken into account to draw a reliable conclusion in future studies.

Primary knee osteoarthritis remains a difficult-to-control degenerative disease. With the rise in average life expectancy and the incidence of obesity, osteoarthritis has brought an increasing economic and physical burden on people. This article summarizes the latest understanding of platelet-rich plasma in the treatment of knee osteoarthritis, and reviews the economic issues of PRP.The literatures in Pubmed, Embase, Cochrane library, Web-science and other databases were searched, and literature inclusion and exclusion criteria were formulated. According to the Cochrane systematic reviewer's manual, the included literatures were grouped, and qualitative descriptions and quantitative meta-analysis were performed. Continuous statistical methods were used to compare the effects and adverse effects of PRP before and after treatment, as well as between PRP and other conservative treatments.A total of 12 randomized controlled trials were included in this study. A total of 959 KOA patients (1070 knees) were enrolled and followed for 3-12 months. PRP total knee scores were significantly better than baseline at 1, 2, 3, 6 and 12 months after treatment (1 month: SMD = 0.60, P 0.01; 2 months: SMD = 0.98, P 0.01; 3 months: SMD = 1.16, P 0.01; 6 months: SMD = 1.49, P 0.01; 12 months: SMD = 1.47, P 0.01). In terms of adverse reactions, PRP did not increase the risk of adverse events compared with HA (OR = 0.96, P = 0.85).Compared with many other treatment methods, intra-articular injection of PRP has been proven to be safe and effective to improve the quality of life of patients with KOA.

Aims: Metabolic disorders may play key roles in oxidative stress and neuronal apoptosis in response to early brain injury (EBI) after subarachnoid hemorrhage (SAH). Pyruvate dehydrogenase (PDH) is ...

The key to hyperspectral anomaly detection is to effectively distinguish anomalies from the background, especially in the case that background is complex and anomalies are weak. Hyperspectral imagery (HSI) as an image-spectrum merging cube data can be intrinsically represented as a third-order tensor that integrates spectral information and spatial information. In this article, a prior-based tensor approximation (PTA) is proposed for hyperspectral anomaly detection, in which HSI is decomposed into a background tensor and an anomaly tensor. In the background tensor, a low-rank prior is incorporated into spectral dimension by truncated nuclear norm regularization, and a piecewise-smooth prior on spatial dimension can be embedded by a linear total variation-norm regularization. For anomaly tensor, it is unfolded along spectral dimension coupled with spatial group sparse prior that can be represented by the l2,1-norm regularization. In the designed method, all the priors are integrated into a unified convex framework, and the anomalies can be finally determined by the anomaly tensor. Experimental results validated on several real hyperspectral data sets demonstrate that the proposed algorithm outperforms some state-of-the-art anomaly detection methods.

This paper proposes a novel online and self-learning algorithm to the identification of fuzzy neural networks, which not only learns the structure and parameters online but also learns the threshold parameters by itself and automatically. For structure learning, a self-constructing approach including adding neurons and merging highly similar fuzzy rules is proposed based on the criteria of the system error between actual and model output and the maximum firing strength of neurons. In order to achieve the efficient merging computing, a new calculation method of similarity degree between fuzzy rules is developed. Further and more importantly, the varying width of Gaussian membership functions can be learned by itself according to the underfitting and overfitting criteria. Similarly, different from the existing constant threshold of similarity degree for merging, the varying threshold of similarity degree can be self-learned according to the real-time accuracy of model. The proposed self-learning mechanism significantly improves the model accuracy and greatly enhances the easy usability. Several benchmark examples are implemented to illustrate the effectiveness and feasible of the proposed approach.

Objective/Background Thoracic outlet syndrome (TOS) is most often referred to vascular surgeons. However, there is a lack of understanding of the malpractice cases involving TOS. The goal of this study is to better understand the medicolegal landscape related to the care of TOS. Methods The Westlaw Edge AI-powered proprietary system was retrospectively reviewed for malpractice cases involving TOS. A Boolean search strategy was used to identify target cases under the case category of “Jury Verdicts & Settlements” for all state and federal jurisdictions from 1970 to September 2020. The settled case was described but not included in the statistical analysis. Descriptive statistics were used to report our findings, and when appropriate. The P ≤ .05 decision rule was established a priori as the null hypothesis rejection criterion to determine associations between jury verdicts outcomes and state's tort reform status. Results In this study, 39 cases were identified and met the study's inclusion criteria from the entire Westlaw Edge database. Among plaintiffs who disclosed age and/or gender, median age was 35.0 years with a female majority (67.6%). Cases involving TOS were noted to be steadily decreasing since the mid-1990s. The cases were unevenly spread across 18 states, with the highest number of cases (14, 35.9%) from California and the second highest (4, 10.3%) from Pennsylvania. A similar uneven distribution was seen among U.S. census regions, in which the West had the highest cases (39.5%). The study revealed that more cases were brought to trials in tort reform states (26, 68.4%) than in non-tort reform states (12, 31.6%). A total of 24 of 39 (61.5%) plaintiffs had one specific claim, which resulted in their economic and noneconomic damages. Negligent operation and treatment complication represented an overwhelming majority of claims brought by 38 of 39 plaintiffs (97.4%). Misdiagnosis and lack of informed consent were both brought nine times (23.1%) by the group. Intraoperative nerve injury (20 patients, 51.3%) was the most commonly reported complication. Excluding one case with a settlement of $965,000, 30 of 38 (78.9%) cases went to trials and received defense verdicts. Eight cases (20.5%) were found in favor of plaintiffs with a median payout of $725,581. Conclusions This study highlighted higher than average payouts to plaintiffs and risk factors that may result in malpractice lawsuits for surgeons undertaking TOS treatment. Future studies are needed to further clarify the relationships between tort reform and outcomes of malpractice cases involving TOS.

Pavement potholes have low detection accuracy under the condition of small samples. To address this issue, we propose a method for efficient and accurate pothole detection under small-sample conditions based on an improved Faster R-CNN (Region-based Convolution Neural Networks). First, images consisting of different pothole shapes and sizes were acquired from different sources and then augmented and denoised to obtain an image set. Second, two representative target-detection models, Faster R-CNN and YOLOv3, were tested. The detection results indicate that Faster R-CNN achieves better detection performance. Furthermore, to overcome inconsistencies (missed detections and inaccurate position estimations), the feature extraction layers of VGG16, ZFNet, and ResNet50 networks were used in combination with Faster R-CNN. The results showed that the VGG16+Faster R-CNN fusion model yielded superior accuracy. Finally, the detection accuracy improved to 0.8997 after adjusting the size of the candidate frame, which also enabled successful detection of previously missed targets.

Despite the continuously increased requirement on automated synthesis of medicines for distributed manufacturing and personal care, it remains a challenge to realize automated synthesis which requires solid-liquid phase reactions. In this work, we demonstrated an automated solid-liquid synthesis for gadopentetate dimeglumine, the most widely used magnetic resonance imaging (MRI) contrast agent. The high-efficiency reaction was performed in a 3D microfluidic chip which was fabricated by femtosecond laser micromachining. The structure of the chip realized 3D shear flow which was essential for highly efficient mixing and movement of the solid-liquid mixtures. Ultraviolet visible (UV–vis) spectrometer was employed for in-line analysis to help automation of this system. Comparing with the round-bottom flask system, this synthetic system showed significantly higher reaction rate, indicating the advantage of the 3D microfluidic technology in micro chemical engineering.

The demand of sharing video streaming extremely increases due to the proliferation of Internet of Things (IoT) devices in recent years, and the explosive development of Artificial Intelligent (AI) detection techniques have made visual privacy protection more urgent and difficult than ever before. Although a number of approaches have been proposed, their essential drawbacks limit the effect of visual privacy protection in real applications. In this paper, we propose a cycle Vector Quantized Variational Autoencoder (cycle-VQ-VAE) framework to encode and decode the video with its extracted audio, which takes advantage of multiple heterogeneous data sources in the video itself to protect individuals’ privacy. In our cycle-VQ-VAE framework, a fusion mechanism is designed to integrate the video and its extracted audio. Particularly, the extracted audio works as the random noise with a non-patterned distribution, which outperforms the noise that follows a patterned distribution for hiding visual information in the video. Under this framework, we design two models, including frame-to-frame (F2F) model and video-to-video (V2V) model, to obtain privacy-preserving video streaming. In F2F, the video is processed as a sequence of frames; while, in V2V, the relations between frames are utilized to deal with the video, greatly improving the performance of privacy protection, video compression, and video reconstruction. Moreover, the video streaming is compressed in our encoding process, which can resist side-channel inference attacks during video transmission and reduce video transmission time. Through the real-data experiments, we validate the superiority of our models (F2F and V2V) over the existing methods in visual privacy protection, visual quality preservation, and video transmission efficiency. Our codes and more data are now available in.

Objectives We sought to identify patients with repaired tetralogy of Fallot (rTOF) at high risk of death and malignant ventricular arrhythmia (VA). Background To date there is no robust ri...

In this work, a new method for hand character gesture recognition based on a single millimetre‐wave radar chip is proposed. Most current radar‐based gesture recognition approaches use Doppler spectrograms or the feature extraction of Doppler spectrograms. In contrast, the authors propose using the trajectory points of gesture movements. Compared with the previous methods, the variations in the gesture movement speed, distance and direction during data acquisition have relatively little impact on the generated trajectories. The character trajectories drawn by gestures in the air output a fixed shape, and this fixed output facilitates the classification of gestures at a later stage. In this work, a single millimetre‐wave radar is used to collect data and a three‐dimensional fast Fourier Transform (3D‐FFT) algorithm is used to obtain the gesture trajectories. For the trajectory points, a trajectory conversion image method is proposed and finally the improved Xception network proposed in this work is used for image recognition. Compared with traditional methods, the proposed method does not require large‐scale data acquisition to facilitate different users. This work is concluded by recognising hand character gestures and comparing them with other commonly used classification methods to verify the gesture recognition accuracy of the proposed method.

With the construction of smart cities and the continuous improvement of people's living standards, residents’ demand for fresh agricultural products (FAPs) has increased dramatically. Therefore, reasonable arrangement for intelligent distribution of FAP in smart cities can effectively guarantee product quality, improve distribution efficiency, reduce distribution cost, and increase customer satisfaction. In actual distribution in smart city, road conditions are one of the important factors that affect the distribution. Therefore, according to the influence of road conditions on refrigerated vehicle's (RV's) speed, the RV's speed characteristic models are established. Meanwhile, according to the characteristics of FAP, the penalty cost function based on the time window is constructed. According to the idea of fuzzy logic, the customer satisfaction evaluation model is established. Then, in order to minimize the distribution costs and maximize customer satisfaction as the optimization goal of intelligent distribution in smart city, the mathematical model is built. For solving this model, an improved quantum-behaved particle swarm optimization algorithm (IQPSO) is proposed. Finally, the effectiveness of IQPSO is verified by simulation. The results show that IQPSO also achieves good results, and the model constructed can effectively balance the relationship between the distribution costs and customer satisfaction when distributing FAP in smart city.

We report a photonic approach to generate multi-band phase-coded microwave pulses based on polarization manipulation of optical signals using a dual-polarization dual-parallel Mach-Zehnder modulator (DPol-DPMZM). Two radio frequency signals are sent to the DPol-DPMZM to realize carrier-suppressed double sideband (CS-DSB) modulation. A high-speed polarization modulator, driven by a three-level electrical coding signal s(t), is used to switch the modulation states of optical signals. For s(t)=1 or -1, the phases of the generated microwave signals are 0 or . For s(t)=0, a carrier-suppressed single sideband (CS-SSB) modulation is realized and there is no desired microwave signal recovered after photodetection. In this way, multi-band phase-coded microwave pluses are generated. The phase jump of the generated phase-coded pulses is determined by the polarity of s(t) instead of its amplitude, which is low power consuming and convenient. Compared to most existing photonic schemes that generate multi-band phase-coded signals in continuous wave (CW) mode, our scheme can directly generate multi-band phase-coded signal in pulse mode, which is pretty suitable for multi-band radar system application.

Coopetition is a business phenomenon that dominates many supply chains. We aim to understand why and how coopetition occurs in supply chains with the presence of process improvement in upstream component production. An original equipment manufacturer (OEM) can purchase the component from either a non-competing supplier (NS) or a competing supplier (CS) that also sells substitutable products in the end consumer market. We demonstrate that without process improvement, coopetition does not exist, because the OEM prefers sourcing from the NS, even though the CS is willing to sell the component to the OEM. Two cases are considered in which process improvement is present: supplier-initiated investment, where the suppliers conduct self-investment, and OEM-initiated investment, where the OEM invests in the supplier's process. We find that with supplier-initiated investment, the CS always prefers coopetition, but the OEM may not. When the investment is efficient, coopetition occurs as an equilibrium outcome; otherwise, competition occurs. The cause of coopetition equilibrium comes from variable process improvement incentives of the CS that derive from two driving forces: the vertical wholesale price effect and the horizontal quantity competition effect. When the degree of product substitution is low, the CS increases its investment and lowers its wholesale price, benefiting the OEM. When the degree of product substitution is high, the CS reduces its investment as well as production quantity, which also benefits the OEM. Interestingly, with OEM-initiated investment, the CS might not always choose to sell the component to the OEM, and as a result, coopetition never occurs.

Ta-doped Bi3.25La0.75Ti3O12(BLTT)/ZnO films were fabricated on Pt(111)/Ti/SiO2/Si substrates by a magnetron sputtering method. Firstly, ZnO crystal thin films were grown on the substrates by a reactive sputtering method. Then, BLTT thin films were deposited on the ZnO layers at room temperature and post-annealed at 600 °C. The micromorphology, ferroelectric and dielectric properties of BLTT/ZnO films were analyzed. The XRD analysis shows that ZnO buffer layer significantly reduces the crystallization temperature of BLTT thin film. The TEM results show that lamellar BLTT grains are grown on ZnO layer at a certain angle with few elements diffusion at the interface of ZnO phase and Bi4Ti3O12 phase. The ferroelectric properties indicate that BLTT/ZnO films exhibit different remanent polarization and coercive fields under electric field with different directions. The novel mechanism of tailoring ferroelectric properties may open new possibilities for designing special ferroelectric devices.

Resistive random-access memory (RRAM) is a promising technology to develop nonvolatile memory and artificial synaptic devices for brain-inspired neuromorphic computing. Here, we have developed a STO:Ag/SiO2 bilayer based memristor that has exhibited a filamentary resistive switching with stable endurance and long-term data retention ability. The memristor also exhibits a tunable resistance modulation under positive and negative pulse trains, which could fully mimic the potentiation and depression behavior like a bio-synapse. Several synaptic plasticity functions, including long-term potentiation (LTP) and long-term depression (LTD), paired-pulsed facilitation (PPF), spike-rate-dependent-plasticity (SRDP), and post-tetanic potentiation (PTP), are faithfully implemented with the fabricated memristor. Moreover, to demonstrate the feasibility of our memristor synapse for neuromorphic applications, spike-time-dependent plasticity (STDP) is also investigated. Based on conductive atomic force microscopy observations and electrical transport model analyses, it can be concluded that it is the controlled formation and rupture of Ag filaments that are responsible for the resistive switching while exhibiting a switching ratio of ~103 along with a good endurance and stability suitable for nonvolatile memory applications. Before fully electroforming, the gradual conductance modulation of Ag/STO:Ag/SiO2/p++-Si memristor can be realized, and the working mechanism could be explained by the succeeding growth and contraction of Ag filaments promoted by a redox reaction. This newly fabricated memristor may enable the development of nonvolatile memory and realize controllable resistance/weight modulation when applied as an artificial synapse for neuromorphic computing.

One of the challenges in hyperspectral image (HSI) classification is that there are limited labeled samples to train a classifier for very high-dimensional data. In practical applications, we often encounter an HSI domain (called target domain) with very few labeled data, while another HSI domain (called source domain) may have enough labeled data. Classes between the two domains may not be the same. This article attempts to use source class data to help classify the target classes, including the same and new unseen classes. To address this classification paradigm, a meta-learning paradigm for few-shot learning (FSL) is usually adopted. However, existing FSL methods do not account for domain shift between source and target domain. To solve the FSL problem under domain shift, a novel deep cross-domain few-shot learning (DCFSL) method is proposed. For the first time, DCFSL tackles FSL and domain adaptation issues in a unified framework. Specifically, a conditional adversarial domain adaptation strategy is utilized to overcome domain shift, which can achieve domain distribution alignment. In addition, FSL is executed in source and target classes at the same time, which can not only discover transferable knowledge in the source classes but also learn a discriminative embedding model to the target classes. Experiments conducted on four public HSI data sets demonstrate that DCFSL outperforms the existing FSL methods and deep learning methods for HSI classification. Our source code is available at https://github.com/Li-ZK/DCFSL-2021.

Nowadays, due to the ability of extracting deep features, the deep learning-based anomaly detection (AD) methods for hyperspectral images (HSIs) have been widely studied. However, all these AD methods treat the tasks of feature extraction and AD separately. Besides, most of them also do not make use of abundant spatial information of HSIs. Thus, a spectral-spatial hyperspectral AD method via collaborative representation constraint stacked autoencoders (SSCRSAE) is proposed. First, the collaborative representation constraint is imposed on the stacked autoencoders to extract deep nonlinear features that are more suitable for the collaborative representation-based detector (CRD). Then, CRD is used to for obtaining the preliminary detection result, which is more convenient for real HSIs because of no need for assuming the distribution of the background. Finally, aiming at further improving the SSCRSAE detector's performance, a novel spectral-spatial AD procedure is designed for calculating the final detection result by considering the spatial information of an HSI. Experimental results express that the proposed SSCRSAE exceeds eight state-of-the-art anomaly detectors used for comparison.

The human imperceptible adversarial examples crafted by l 0 -norm attacks, which aims to minimize l 0 distance from the original image, thereby misleading deep neural network classifiers into the wrong classification. Prior works of tackling l 0 attacks can neither eliminate perturbed pixels nor improve the performance of the classifier in the recovered low-quality images. To address the issue, we propose a novel method, called space transformation pixel defender (STPD), to transform any image into a latent space to separate the perturbed pixels from the normal pixels. In particular, this strategy uses a set of one-class classifiers, including Isolation Forest and Elliptic Envelope, to locate the perturbed pixels from adversarial examples. The value of the neighboring normal pixels is then used to replace the perturbed pixels, which hold more than half of the votes from these one-class classifiers. We use our proposed strategy to successfully defend against well-known l 0 -norm adversarial examples in the image classification settings. We show experimental results under the One-pixel Attack (OPA), the Jacobian-based Saliency Map Attack (JSMA), and the Carlini Wagner (CW) l 0 -norm attack on CIFAR-10, COVID-CT, and ImageNet datasets. Our experimental results show that our approach can effectively defend against l 0 -norm attacks compared with the most popular defense techniques.

Hyperspectral image (HSI) has hundreds of continuous bands that contain a lot of redundant information. Besides, a spatial patch of a hyperspectral cube often contains some pixels different from the center pixel category, which are usually called interference pixels. The existence of such interference pixels has a negative effect on extracting more discriminative information. Therefore, in this letter, a multiattention fusion network (MAFN) for HSI classification is proposed. Compared with the current state-of-the-art methods, MAFN uses band attention module (BAM) and spatial attention module (SAM), respectively, to alleviate the influence of redundant bands and interfering pixels. In this way, MAFN realizes feature reuse and obtains complementary information from different levels by combining multiattention and multilevel fusion mechanisms, which can extract more representative features. Experiments were conducted on two public HSI data sets to demonstrate the effectiveness of MAFN. Our source code is available at https://github.com/Li-ZK/MAFN-2021.

To overcome the limitations of pavement skid resistance prediction using the friction coefficient, a genetic-algorithm-improved neural network (GAI-NN) was developed in this study. First, three-dimensional (3D) point-cloud data of an asphalt pavement surface were obtained using a smart sensor (Gocator 3110). The friction coefficient of the pavement was then obtained using a pendulum friction tester. The 3D point-cloud dataset was then analyzed to recover the missing data and perform denoising. In particular, these data were filled using cubic spline interpolation. Parameters for texture characterization were defined, and methods for computing the parameters were developed. Finally, the GAI-NN model was developed by modifying the weights and thresholds. The test results indicated that using pavement surface texture 3D data, the GAI-NN was capable of predicting the pavement friction coefficient with sufficient accuracy, with an error of 12.1%.

Although solar energy is promising for water purification, there is still a room for further improving the solar steam generation efficiency. Herein, an environmental energy-enhanced solar steam evaporator is fabricated by immersing a cellulose acetate fiber-based cigarette filter (CF) in an aqueous solution of polyvinyl alcohol (PVA) followed by freeze-drying and decorating with MXene sheets. The presence of MXene is to absorb solar light and convert solar energy to thermal energy for efficient water evaporation, while the porous PVA network generated inside the pores of the filter during the freeze-drying process accommodates the dispersed MXene sheets and interconnects the CF and MXene. Because of the constructed PVA/MXene network inside the CF porous architecture and the hydrophilic feature of both MXene and PVA, the resultant MXene/PVA modified CF (MPCF) is highly hydrophilic and competent for rapid upward transfer of water. Interestingly, in addition to the normal energy input by the incident solar light, the large-area sidewall of MPCF gains thermal energy from the environment in the forms of heat convection and heat radiation to enhance the solar steam generation efficiency, resulting in an ultrahigh water evaporation rate of 3.38 kg m−2 h−1 with an outstanding evaporation efficiency of 132.9%.

Objectives: To investigate the clinical distribution and drug resistance of Klebsiella pneumoniae pulmonary infection in patients with hypertensive intracerebral hemorrhage after minimally invasive surgery. Methods: A total of 658 patients with hypertensive intracerebral hemorrhage who underwent minimally invasive surgery admitted to the intensive care unit (ICU) and the Department of Neurology of Affiliated Hospital of Hebei University from January 2015 to January 2020 were enrolled and divided into two groups: the observation group and the control group. Three hundred and thirty-three cases with postoperative pulmonary infection were included into the observation group, and 325 cases without postoperative pulmonary infection were divided into the control group. The intubation time, neurological deficiency score and Glasgow coma scale (GCS) of the two groups were analyzed and compared. Automatic microbial identification system was utilized to isolate bacteria from patients in the observation group, identify Klebsiella pneumoniae, and analyze Klebsiella pneumoniae infection, clinical department distribution, and age distribution. The Kirby-Bauer method was adopted to carry out the drug susceptibility test of Klebsiella pneumoniae infection. Results: The intubation time and neurological deficiency score of patients with hypertensive cerebral hemorrhage in the observation group were significantly higher than those in the control group (p

Hyperspectral image (HSI) classification has drawn increasing attention recently. However, it suffers from noisy labels that may occur during field surveys due to a lack of prior information or human mistakes. To address this issue, this article proposes a novel dual-channel residual network (DCRN) to resolve HSI classification with noisy labels. Currently, the influence of noisy labels is reduced by simply detecting and removing those anomalous samples. Different from such a specifically designed noise cleansing method, DCRN is easy to implement but highly effective. It enhances its model robustness to noisy labels to a great extent by employing a novel dual-channel structure and a noise-robust loss function. In this way, DCRN can mitigate influence from noisy labels while fully utilizing useful information from mislabeled samples for augmented training. Experiments are conducted on several hyperspectral data sets with manually generated noisy labels to demonstrate its excellent performance. The code is available at https://github.com/Li-ZK/DCRN-2021.

Background Systemic lupus erythematosus (SLE) is a chronic and complex multi-system autoimmune disorder. Higher risks of hematological malignancies (HM) were observed in SLE patients, which was associated with higher mortality. The mechanism and risk factors of HM oncogenesis in SLE patients are still under investigation. The aim of this study was to explore clinical characteristics, risk factors, and prognosis of SLE patients with or without HM in the Chinese population. Methods A retrospective, case-controlled study was conducted in 72 SLE patients between January 2013 and December 2020. Clinical and laboratory data were collected and compared between the two groups of patients with HM and those without HM. Logistic regression analysis was performed to determine risk factors of HM oncogenesis. The survival rate was estimated by Kaplan-Meier methods and Cox proportional hazards regression analysis. Results Among 72 SLE patients in this study, fifteen complicated with HM and 57 without HM were identified. The incidence rate of HM was approximately 0.24% with elevated standardized incidence ratios of lymphoma and leukemia (27.559 and 12.708, respectively). Patients with HM were older when diagnosed with SLE, with a higher frequency of infection and splenomegaly, lower levels of hemoglobin and high-density lipoprotein compared with those without HM. Fewer patients with HM expressed positive anti-dsDNA antibody (26.7% vs 66.7%, P = 0.005) or received hydroxychloroquine treatment (40.0% vs 86.0%, P = 0.001). Older age at SLE diagnosis (OR=1.122, 95% CI: 1.037–1.214) was regarded as an independent risk factor of HM oncogenesis. Female (RR= 0.219, 95% CI: 0.070–0.681) and hydroxychloroquine (RR= 0.281, 95% CI: 0.094–0.845) were protective factors of mortality in SLE patients. Conclusions SLE patients with an older age are at an increased risk of HM carcinogenesis. The prognosis of male patients with SLE tends to be poorer whether complicated with HM. The association of antinuclear antibody spectrum, medication, and HM oncogenesis in SLE needs further investigation.