Search

IntroductionThe association between a plant-based diet and weight loss and atherosclerotic cardiovascular disease (ASCVD) has not been fully elucidated. We performed a pooled analysis and Mendelian randomization (MR) analysis to investigate this question.MethodsWe searched for randomized controlled trials on the effects of a plant-based diet on weight loss compared with a non-plant-based diet. In addition, a two-sample MR study was conducted. IVs were obtained from the genome-wide association studies (GWAS) on the exposures, and we obtained summary statistics on the outcomes. The inverse-variance weighted (IVW) method was used as the main analysis and other MR methods were performed as supplementary analyses.ResultsIndividuals on the plant-based diet lost more weight than the non-plant-based diet group (WMD −0.96 kg; 95% CI: −1.32 to −0.60). Population conditions and energy restriction were identified as the study-level factors that influenced the pooling results in the subgroup analyses. Increased consumption of raw vegetables was significantly associated with lower BMI (IVW, β −0.35, 95% CI: −0.62 to −0.08, p = 0.012) and lower risk of obesity (IVW, OR 0.11, 95% CI: 0.01 to 0.99, p = 0.048), coronary heart disease (IVW, OR 0.44, 95% CI: 0.21 to 0.92, p = 0.029) and myocardial infarction (IVW, OR 0.39,95% CI: 0.15 to 0.98, p = 0.045) and a higher HDL-C (IVW, β 0.47, 95% CI: 0.24 to 0.70, p = 4×10−5).DiscussionThe present findings suggest that raw vegetable intake is beneficial for weight loss and prevention of ASCVD.

With the increasing integration of distributed photovoltaic (PV) generation into distribution networks, challenges such as power reverse flow and high line losses have emerged, leading to greater uncertainty in power systems. To address these issues, this paper presents an analytical model for calculating line losses in low-voltage distribution networks with PV generation, utilizing power flow calculations. A simulation model of a 15 node low-voltage network is developed using SIMULINK to validate the accuracy of the analytical model under the scenario of uniform load distribution (ULD). Additionally, a line loss optimization algorithm based on quantum genetic algorithms (QGA) is proposed for low-voltage distribution networks with distributed PV generation, along with an optimization model. The objective function of the optimization model is based on the reduction in line losses resulting from the integration of the PV system. The example results demonstrate the consistency between the line loss optimization using QGA and the analytical results, highlighting the significant advantages of QGA in terms of speed and accuracy. This research provides valuable insights for line loss optimization in low-voltage distribution networks with distributed PV generation and serves as a theoretical reference for future studies in this field.

Abstract Background Universal stress proteins (USPs) are a class of stress-induced proteins that play a crucial role in biotic and abiotic stress responses. These proteins have previously been reported to participate directly in responses to various stress and protect plants against unfavorable environmental conditions. However, there is limited research on USPs in cotton, and systematic characterization of USPs in Gossypium species is lacking. Results In the present study, the USP genes in Gossypium hirsutum were systematically identified and classified into six distinct subfamilies. The expansion of USPs in Gossypium species is mainly caused by dispersed duplication and whole genome duplication. Notably, the USPs that have expanded through allotetraploidization events are highly conserved in the allotetraploid species. The promoter regions of GhUSPs contain a diverse range of cis-acting elements associated with stress response. The RNA-Seq analysis and RT-qPCR assays revealed a significant induction of numerous GhUSPs expressions in response to various abiotic stresses. The co-expression network of GhUSPs revealed their involvement in stress response. Conclusions This study systematically analyzed the biological characteristics of GhUSPs and their response to abiotic stress. These findings serve as a theoretical basis for facilitating the breeding of cotton varieties in future research.

Aerial image target detection technology has essential application value in navigation security, traffic control and environmental monitoring. Compared with natural scene images, the background of aerial images is more complex, and there are more small targets, which puts higher requirements on the detection accuracy and real-time performance of the algorithm. To further improve the detection accuracy of lightweight networks for small targets in aerial images, we propose a cross-scale multi-feature fusion target detection method (CMF-YOLOv5s) for aerial images. Based on the original YOLOv5s, a bidirectional cross-scale feature fusion sub-network (BsNet) is constructed, using a newly designed multi-scale fusion module (MFF) and cross-scale feature fusion strategy to enhance the algorithm's ability, that fuses multi-scale feature information and reduces the loss of small target feature information. To improve the problem of the high leakage detection rate of small targets in aerial images, we constructed a multi-scale detection head containing four outputs to improve the network's ability to perceive small targets. To enhance the network's recognition rate of small target samples, we improve the K-means algorithm by introducing a genetic algorithm to optimize the prediction frame size to generate anchor boxes more suitable for aerial images. The experimental results show that on the aerial image small target dataset VisDrone-2019, the proposed method can detect more small targets in aerial images with complex backgrounds. With a detection speed of 116 FPS, compared with the original algorithm, the detection accuracy metrics mAP0.5 and mAP0.5:0.95 for small targets are improved by 5.5% and 3.6%, respectively. Meanwhile, compared with eight advanced lightweight networks such as YOLOv7-Tiny and PP-PicoDet-s, mAP0.5 improves by more than 3.3%, and mAP0.5:0.95 improves by more than 1.9%.

Chestnut is often used in cloudy beverage for its good nutritional value and health care function. Due to abundant starch in chestnut, the chestnut beverage is easy to precipitate or stratify. Therefore, in this study, microcrystalline cellulose (MCC) and agar were composited to improve the stability of the chestnut suspension. The determination of particle size, viscosity, rheology and Turbiscan were carried out to evaluate the effect of MCC concentration on the stability of chestnut suspension. The results showed that the chestnut suspension was relatively stable with smaller particle size and higher potential value when MCC concentration was 2.4 g/L and agar concentration was 0.4 g/L, but little change in particle size occurred during thermal acceleration process. When MCC concentration was 2.4 and 2.8 g/L, the chestnut suspension had higher viscosity, and could provide better suspension stability for suspended particles in the system. The rheological result showed that when the concentration of MCC was 2.4 g/L, a relatively stable network structure could be formed in the system, resulting in the stronger damage resistance. The Turbiscan analysis showed that the TSI value of the system was relatively small, leading to stable within a short time when MCC concentration was 2.4 and 2.8 g/L.

Electroencephalography (EEG) signals are gaining popularity in Brain-Computer Interface (BCI)-based rehabilitation and neural engineering applications thanks to their portability and availability. Inevitably, the sensory electrodes on the entire scalp would collect signals irrelevant to the particular BCI task, increasing the risks of overfitting in machine learning-based predictions. While this issue is being addressed by scaling up the EEG datasets and handcrafting the complex predictive models, this also leads to increased computation costs. Moreover, the model trained for one set of subjects cannot easily be adapted to other sets due to inter-subject variability, which creates even higher over-fitting risks. Meanwhile, despite previous studies using either convolutional neural networks (CNNs) or graph neural networks (GNNs) to determine spatial correlations between brain regions, they fail to capture brain functional connectivity beyond physical proximity. To this end, we propose 1) removing task-irrelevant noises instead of merely complicating models; 2) extracting subject-invariant discriminative EEG encodings, by taking functional connectivity into account. Specifically, we construct a task-adaptive graph representation of the brain network based on topological functional connectivity rather than distance-based connections. Further, non-contributory EEG channels are excluded by selecting only functional regions relevant to the corresponding intention. We empirically show that the proposed approach outperforms the state-of-the-art, with around 1% and 11% improvements over CNN-based and GNN-based models, on performing motor imagery predictions. Also, the task-adaptive channel selection demonstrates similar predictive performance with only 20% of raw EEG data, suggesting a possible shift in direction for future works other than simply scaling up the model.

Abstract Background Older adult patients mainly suffer from multiple comorbidities and are at a higher risk of deep venous thrombosis (DVT) during their stay in the intensive care unit (ICU) than younger adult patients. This study aimed to analyze the risk factors for DVT in critically ill older adult patients. Methods This was a subgroup analysis of a prospective, multicenter, observational study of patients who were admitted to the ICU of 54 hospitals in Zhejiang Province from September 2019 to January 2020 (ChiCTR1900024956). Patients aged > 60 years old on ICU admission were included. The primary outcome was DVT during the ICU stay. The secondary outcomes were the 28- and 60-day survival rates, duration of stay in ICU, length of hospitalization, pulmonary embolism, incidence of bleeding events, and 60-day coagulopathy. Results A total of 650 patients were finally included. DVT occurred in 44 (2.3%) patients. The multivariable logistic regression analysis showed that age (≥75 vs 60-74 years old, odds ratio (OR) = 2.091, 95% confidence interval (CI): 1.308-2.846, P = 0.001), the use of analgesic/sedative/muscarinic drugs (OR = 2.451, 95%CI: 1.814-7.385, P = 0.011), D-dimer level (OR = 1.937, 95%CI: 1.511-3.063, P = 0.006), high Caprini risk score (OR = 2.862, 95%CI: 1.321-2.318, P = 0.039), basic prophylaxis (OR = 0.111, 95%CI: 0.029-0.430, P = 0.001), and physical prophylaxis (OR = 0.322, 95%CI: 0.109-0.954, P = 0.041) were independently associated with DVT. There were no significant differences in 28- and 60-day survival rates, duration of stay in ICU, total length of hospitalization, 60-day pulmonary embolism, and coagulation dysfunction between the two groups, while the DVT group had a higher incidence of bleeding events (2.6% vs. 8.9%, P

The use of artificial intelligence (AI) to generate automated early warnings in epidemic surveillance by harnessing vast open-source data with minimal human intervention has the potential to be both revolutionary and highly sustainable. AI can overcome the challenges faced by weak health systems by detecting epidemic signals much earlier than traditional surveillance. AI-based digital surveillance is an adjunct to—not a replacement of—traditional surveillance and can trigger early investigation, diagnostics and responses at the regional level. This narrative review focuses on the role of AI in epidemic surveillance and summarises several current epidemic intelligence systems including ProMED-mail, HealthMap, Epidemic Intelligence from Open Sources, BlueDot, Metabiota, the Global Biosurveillance Portal, Epitweetr and EPIWATCH. Not all of these systems are AI-based, and some are only accessible to paid users. Most systems have large volumes of unfiltered data; only a few can sort and filter data to provide users with curated intelligence. However, uptake of these systems by public health authorities, who have been slower to embrace AI than their clinical counterparts, is low. The widespread adoption of digital open-source surveillance and AI technology is needed for the prevention of serious epidemics.

Diagnosing the fault type accurately from a variety of faults is very essential to ensure a stable electricity supply when a short-circuit fault occurs. In this paper, a hybrid classification model combining the one-dimensional convolutional neural network (1D-CNN) and the bidirectional long short-term memory network (BiLSTM) is proposed for the classification of cable short-circuit faults to improve the accuracy of fault diagnosis. Sample sets of the current signal for single-phase grounding short circuit, two-phase grounding short circuit, two-phase to phase short circuit, and three-phase grounding short-circuit are obtained by the simulink model, and the signal is input to this network model. The local features of the cable fault signals are extracted using 1D-CNN and the fault signal timing information is captured using BiLSTM, which enables the diagnosis of cable faults based on the automatically extracted features. The experimental results of the simulation show that the model can obtain a good recognition performance and can achieve an overall accuracy of 99.45% in classifying the four short-circuit faults with 500 iterations. In addition, the analysis of loss function curves and accuracy curves shows that the method performs better than networks with only temporal feature extraction, such as 1D-CNN and LSTM.

With the development of digital technology, machine learning has paved the way for the next generation of tinnitus diagnoses. Although machine learning has been widely applied in EEG-based tinnitus analysis, most current models are dataset-specific. Each dataset may be limited to a specific range of symptoms, overall disease severity, and demographic attributes; further, dataset formats may differ, impacting model performance. This paper proposes a side-aware meta-learning for cross-dataset tinnitus diagnosis, which can effectively classify tinnitus in subjects of divergent ages and genders from different data collection processes. Owing to the superiority of meta-learning, our method does not rely on large-scale datasets like conventional deep learning models. Moreover, we design a subject-specific training process to assist the model in fitting the data pattern of different patients or healthy people. Our method achieves a high accuracy of 73.8% in the cross-dataset classification. We conduct an extensive analysis to show the effectiveness of side information of ears in enhancing model performance and side-aware meta-learning in improving the quality of the learned features.

Bendamustine (BENDA) is a bifunctional alkylating agent with alkylating and purinergic antitumor activity, which exerts its anticancer effects by direct binding to DNA, but the detailed mechanism of BENDA–DNA interaction is poorly understood. In this paper, the interaction properties of the anticancer drug BENDA with calf thymus DNA (ctDNA) were systematically investigated based on surface-enhanced Raman spectroscopy (SERS) technique mainly using a novel homemade AuNPs/ZnCl2/NpAA (NpAA: nano porous anodic alumina) solid-state substrate and combined with ultraviolet–visible spectroscopy and molecular docking simulation to reveal the mechanism of their interactions. We experimentally compared and studied the SERS spectra of ctDNA, BENDA, and BENDA–ctDNA complexes with different molar concentrations (1:1, 2:1, 3:1), and summarized their important characteristic peak positions, their peak position differences, and hyperchromic/hypochromic effects. The results showed that the binding modes include covalent binding and hydrogen bonding, and the binding site of BENDA to DNA molecules is mainly the N7 atom of G base. The results of this study help to understand and elucidate the mechanism of BENDA at the single-molecule level, and provide guidance for the further development of effective new drugs with low toxicity and side effects.

Adversarial attacks, e.g., adversarial perturbations of the input and adversarial samples, pose significant challenges to machine learning and deep learning techniques, including interactive recommendation systems. The latent embedding space of those techniques makes adversarial attacks challenging to detect at an early stage. Recent advance in causality shows that counterfactual can also be considered one of the ways to generate the adversarial samples drawn from different distribution as the training samples. We propose to explore adversarial examples and attack agnostic detection on reinforcement learning (RL)-based interactive recommendation systems. We first craft different types of adversarial examples by adding perturbations to the input and intervening on the casual factors. Then, we augment recommendation systems by detecting potential attacks with a deep learning-based classifier based on the crafted data. Finally, we study the attack strength and frequency of adversarial examples and evaluate our model on standard datasets with multiple crafting methods. Our extensive experiments show that most adversarial attacks are effective, and both attack strength and attack frequency impact the attack performance. The strategically-timed attack achieves comparative attack performance with only 1/3 to 1/2 attack frequency. Besides, our white-box detector trained with one crafting method has the generalization ability over several other crafting methods.

Abstract Herein, a human identification system for smart spaces called Vein‐ID (referred to as VID) is presented, which leverage the uniqueness of vein patterns embedded in dorsum of an individual's hand. VID extracts vein patterns using the depth information and infrared (IR) images, both obtained from a commodity depth camera. Two deep learning models (CNN and Stacked‐Autoencoders) are presented for precisely identifying a target individual from a set of N enrolled users. VID also incorporates a strategy for identifying an intruder—that is a person whose vein patterns are not included in the set of enrolled individuals. The performance of VID by collecting a comprehensive data set of approximately 17,500 images from 35 subjects is evaluated. The tests reveal that VID can identify an individual with an average accuracy of over 99% from a group of up to 35 individuals. It is demonstrated that VID can detect intruders with an average accuracy of about 96%. The execution time for training and testing the two deep learning models on different hardware platforms is also investigated and the differences are reported.

Lumbar-pelvic movements (LPMs) are generally performed in the clinical setting to identify limitations in a range of movements. Continuous monitoring of these movements can provide real-time feedback to both patients and medical experts with the potential of identifying activities that may precipitate symptoms of low back pain (LBP) as well as improving therapy by providing a personalised approach. Recent advances in mobile computing technology and wearable sensors have paved the way for developing mobile physical activity monitoring applications with more advanced and complex algorithms, such as deep neural network (DNN) based models. However, there is a lack of prior studies that focus on real-time LPMs detection with multimodal sensory data. Meanwhile, most research in the area of body movement detection do not consider the potential transition logic of the constituent low-level body movements (e.g., LPMs) within their corresponding high-level physical activity. This information could significantly increase accuracy of detection results. Further, current studies mainly perform deep learning-based movement detection in the cloud (or a backend server) that could increase network bandwidth and response time. To address these limitations, this paper proposes a novel LPMs detection approach using an enhanced and adapted hybrid DNN model, which includes a convolutional neural network followed by a long-short term memory recurrent network (CNN-LSTM), and performs detection locally on the mobile device. The results of a comparative evaluation of the proposed model and baseline models are described. We also introduce a set of domain-specific pre-defined rules, based on the transition logic information, to reconstruct the detection outputs to further improve the detection accuracy.

To date, quality-related multivariate statistical methods are extensively used in process monitoring and have achieved admirable effects. However, most of them contain recursive processes, which result in higher time complexity and are not suitable for increasingly complex industrial processes. Therefore, this paper embeds singular value decomposition (SVD) into the kernel principal component regression (KPCR) to accomplish Quality-related process monitoring with a lower computational cost. Specifically, the kernel technique is devoted to map the original input into the higher dimensional space to boost the nonlinear ability of the principal component regression (PCR), and then the KPCR is employed to capture the correlation between the input kernel matrix and the output matrix. At the same time, the kernelized input space is decomposed into two orthogonal quality-related and quality-unrelated spaces by SVD, and the statistics of the two spaces are calculated to detect the faults respectively. Compared with other multivariate statistical methods, it has the following advantages: 1) A quality-related kernel principal component analysis (QR-KPCR) algorithm is proposed. 2) Compared with partial least squares method, the recursive process is omitted and the training time is shortened. 3) The model is more concise and the fault detection process is faster. 4) By contrast with other multivariate statistical process monitoring, it has a higher fault detection rate. Experimental results on a widespread example and an industry benchmark verify the effectiveness and reliability of the proposed method.

AbstractPurpose The aim of this study is to investigate the prevention and treatment patterns of deep vein thrombosis (DVT) in critically ill patients and to explore the risk factors for DVT in people from Zhejiang Province, China.Materials and methods This study prospectively enrolled patients admitted in intensive care units (ICUs) of 54 hospitals from 09/16/2019 to 01/16/2020. The risk of developing DVT and subsequent prophylaxis was evaluated. The primary outcome was DVT occurrence during ICU hospitalisation. Univariate and multivariate logistic regression were performed to determine the risk factors for DVT.Results A total of 940 patients were included in the study. Among 847 patients who received prophylaxis, 635 (75.0%) patients received physical prophylaxis and 199 (23.5%) patients received drug prophylaxis. Fifty-eight (6.2%) patients were diagnosed with DVT after admission to the ICU, and 36 patients were treated with anticoagulants (all patients received low molecular weight heparin [LMWH]). D-dimer levels (OR = 1.256, 95% CI: 1.132–1.990), basic prophylaxis (OR = 0.092, 95% CI: 0.016–0.536), and physical prophylaxis (OR = 0.159, 95% CI: 0.038–0.674) were independently associated with DVT in ICU patients. The short-term survival was similar between DVT and non-DVT patients.Conclusions DVT prophylaxis is widely performed in ICU patients. Prophylaxis is an independent protective factor for DVT occurrence. The most common treatment of DVT patients is LMWH, although it might increase the rate of bleeding.Key messagesThis is the only multicenter and prospective study of DVT in ICUs in China.d-dimer levels were independently associated with DVT in ICU patients.Prophylaxis was an independent protective factor for DVT occurrence in ICU.

The frequency of serious epidemics has continued to increase in the last decade. The ability to predict the risk of outbreaks can improve prevention and control. There are few prediction models available, and of these most are manually constructed by human experts. These manual models are affected by the lack of automation and have limitations in data processing. They can be enhanced with modern machine-learning techniques. Machine learning (ML) based prediction models, however, have higher requirement for professional knowledge and are not broadly accessible to researchers who do not have ML expertise. We proposed automated machine learning (AutoML) as an advanced solution. It automates the entire ML design process for users without requiring ML knowledge, therefore allows non-ML experts to individually build ML models. To demonstrate the functionality of AutoML to develop reliable systems, we expanded an existing manually developed risk analysis model, EPIRISK, that uses economic, social and medical risk factors to predict epidemic risk. Using the AutoML platform BrewAI, we obtained an automatically generated ML model to predict epidemic risk. This was compared with six traditionally built machine learning models. The AutoML tool generated a model of 77% accuracy in predicting risk. It had similar accuracy to the six traditional built ML models. Such tools are easy to use and could make ML models more accessible to non-ML experts.

Under the guidance of the directed graph, this paper studies the fault diagnosis (FD) and fault-tolerant control (FTC) of the leader-follower multi-agent systems (MASs) with time delay. Actuator faults and disturbances are considered in this study, which make the work of this paper more challenging. Firstly, a robust distributed observer on the basis of the relative estimation error is put forward to obtain the fault estimation information. Then, after getting the fault information obtained by the observer, a dynamic compensation FTC protocol on the basis of the relative output tracking error is developed to make the output of followers can track the leader's output. Finally, the proposed algorithm is verified by an example and satisfactory results are obtained.

ZIF-67 is a three-dimensional zeolite imidazole ester framework material with a porous rhombic dodecahedral structure, a large specific surface area and excellent thermal stability. In this paper, the catalytic effect of ZIF-67 on five kinds of energetic materials, including RDX, HMX, CL-20, AP and the new heat-resistant energetic compound DAP-4, was investigated. It was found that when the mass fraction of ZIF-67 was 2%, it showed excellent performance in catalyzing the said compounds. Specifically, ZIF-67 reduced the thermal decomposition peak temperatures of RDX, HMX, CL-20 and DAP-4 by 22.3 °C, 18.8 °C, 4.7 °C and 10.5 °C, respectively. In addition, ZIF-67 lowered the low-temperature and high-temperature thermal decomposition peak temperatures of AP by 27.1 °C and 82.3 °C, respectively. Excitingly, after the addition of ZIF-67, the thermal decomposition temperature of the new heat-resistant high explosive DAP-4 declined by approximately 10.5 °C. In addition, the kinetic parameters of the RDX+ZIF-67, HMX+ZIF-67, CL-20+ZIF-67 and DAP-4+ZIF-67 compounds were analyzed. After the addition of the ZIF-67 catalyst, the activation energy of the four energetic materials decreased, especially HMX+ZIF-67, whose activation energy was approximately 190 kJ·mol−1 lower than that reported previously for HMX. Finally, the catalytic mechanism of ZIF-67 was summarized. ZIF-67 is a potential lead-free, green, insensitive and universal EMOFs-based energetic burning rate catalyst with a bright prospect for application in solid propellants in the future.

When the desired output probability density function (PDF) is unknown, the active fault-tolerant control (FTC) method for the non-Gaussian nonlinear singular stochastic distribution control (SDC) system is investigated in this paper. Algebraic constraints and the nonlinearity in singular systems make the design of fault diagnosis and fault-tolerant control more complex. Different from traditional static modeling methods, the linear fuzzy logic system is served for approximating the output PDF. Takagi-Sugeno (T-S) fuzzy model is used to describe the nonlinear system. Subsequently, a fuzzy descriptor fault diagnosis (FD) observer is used to provide the unknown fault information for the fault-tolerant controller design. Combining minimum entropy control and fault compensation algorithm, the minimum Shannon entropy fault tolerant control strategy is developed to compensate the performance losses caused by the fault. At last, simulation results are applied to demonstrate the effectiveness of the proposed algorithms.

It is of great fundamental significance and practical application to understand the binding sites and dynamic process of the interaction between doxorubicin (DOX) and DNA molecules. Based on the Confocal Raman spectroscopy, the interaction between DOX and calf thymus DNA has been systemically investigated, and some meaningful findings have been found. DOX molecules can not only interact with all four bases of DNA molecules, i.e., adenine, thymine, cytosine, guanine, and phosphate, but also affect the DNA conformation. Meanwhile, the binding site of DOX and its derivatives such as daunorubicin and epirubicin is certain. Furthermore, the interaction between DOX and DNA molecules is a dynamic process since the intensities of each characteristic peaks of the base, e.g., adenine, cytosine, and phosphate, are all regularly changed with the interaction time. Finally, a dynamic mechanism model of the interaction between DOX and DNA molecules is proposed; that is, there are two kinds of interaction between DOX and DNA molecules: DOX-DNA acts to form a complex, and DOX-DOX acts to form a multimer. The two effects are competitive, as the former compresses DNA molecules, and the latter decompresses these DNA molecules. This work is helpful for accurately understanding and developing new drugs and pathways to improve and treat DOX-induced cytotoxicity and cardiotoxicity.

In this paper, a method of fault diagnosis (FD) and fault-tolerant tracking control (FTTC) is investigated for non-Gaussian nonlinear stochastic distribution control (SDC) systems with missing measurements. The phenomenon of the randomly occurring missing measurements is described as a Bernoulli process. The missing measurements during transmission are compensated with the data successfully transmitted at the previous moment. The residual signal of the fault diagnosis observer is different from that of the general system. Using the integral of the error of the output probability density function (PDF) as the driving information, the system state and fault can be estimated by an adaptive fault diagnosis observer. Then, a novel fault-tolerant tracking controller is designed based on a discrete-time 2-order sliding mode to make the post-fault PDF still track the target. Two simulated examples are included to illustrate the effectiveness of the theoretical results.

To deal with the low accuracy of positioning for mobile robots when only using visual sensors and an IMU, a method based on tight coupling and nonlinear optimization is proposed to obtain a high-precision visual positioning scheme by combining measured value of the preintegrated inertial measurement unit (IMU) and values of the odometer and characteristic observations. First, the preprocessing part of the observation data includes tracking of the image data and the odometer data, and preintegration of IMU data. Second, the initialization part of the above three sensors includes IMU preintegration, odometer preintegration, and gyroscope bias calculation. It also includes the alignment of speed, gravity, and scale. Finally, a local BA (bundle adjustment) joint optimization and global graph optimization are established, so as to obtain more accurate positioning results.

Recommendation systems are crucial in the provision of services to the elderly with Alzheimer’s disease in IoT-based smart home environments. In this work, a Reminder Care System (RCS) is presented to help Alzheimer patients live in and operate their homes safely and independently. A contextual bandit approach is utilized in the formulation of the proposed recommendation system to tackle dynamicity in human activities and to construct accurate recommendations that meet user needs without their feedback. The system was evaluated based on three public datasets using a cumulative reward as a metric. Our experimental results demonstrate the feasibility and effectiveness of the proposed Reminder Care System for real-world IoT-based smart home applications.

Lysyl oxidase-like 2 (LOXL2) is an enzyme that promotes scaffolding of extracellular matrix proteins. Here the authors show that LOXL2 is crucial for pressure-overload induced cardiac fibrosis, and that antibody-mediated inhibition or genetic disruption ofLoxl2in mice shows therapeutic potential for treatment of cardiac fibrosis.

In this paper, an integrated scheme including fault diagnosis and fault-tolerant controller design is proposed for the manipulator system with the sensor fault. Any constant fault or time-varying fault can be estimated by the fault diagnosis scheme based on the adaptive observer rapidly and accurately, and the designed parameters can be solved by the linear matrix inequality. Using the fault estimation information, a fault-tolerant controller combining the characteristics of the proportional differentiation control and the sliding model control is designed to trace the expected trajectory via the back-stepping method. Finally, the effectiveness of the above scheme is verified by the simulation results.

Background: A nanomaterial-based electronic-skin (E-Skin) wearable sensor has been successfully used for detecting and measuring body movements such as finger movement and foot pressure. The ultrathin and highly sensitive characteristics of E-Skin sensor make it a suitable alternative for continuously out-of-hospital lumbar−pelvic movement (LPM) monitoring. Monitoring these movements can help medical experts better understand individuals’ low back pain experience. However, there is a lack of prior studies in this research area. Therefore, this paper explores the potential of E-Skin sensors to detect and measure the anatomical angles of lumbar−pelvic movements by building a linear relationship model to compare its performance to clinically validated inertial measurement unit (IMU)-based sensing system (ViMove). Methods: The paper first presents a review and classification of existing wireless sensing technologies for monitoring of body movements, and then it describes a series of experiments performed with E-Skin sensors for detecting five standard LPMs including flexion, extension, pelvic tilt, lateral flexion, and rotation, and measure their anatomical angles. The outputs of both E-Skin and ViMove sensors were recorded during each experiment and further analysed to build the comparative models to evaluate the performance of detecting and measuring LPMs. Results: E-Skin sensor outputs showed a persistently repeating pattern for each movement. Due to the ability to sense minor skin deformation by E-skin sensor, its reaction time in detecting lumbar−pelvic movement is quicker than ViMove by ~1 s. Conclusions: E-Skin sensors offer new capabilities for detecting and measuring lumbar−pelvic movements. They have lower cost compared to commercially available IMU-based systems and their non-invasive highly stretchable characteristic makes them more comfortable for long-term use. These features make them a suitable sensing technology for developing continuous, out-of-hospital real-time monitoring and management systems for individuals with low back pain.

The human gut microbiota impacts host metabolism and has been implicated in the pathophysiology of obesity and metabolic syndromes. However, defining the roles of specific microbial activities and metabolites on host phenotypes has proven challenging due to the complexity of the microbiome-host ecosystem. Here, we identify strains from the abundant gut bacterial phylum Bacteroidetes that display selective bile salt hydrolase (BSH) activity. Using isogenic strains of wild-type and BSH-deleted Bacteroides thetaiotaomicron, we selectively modulated the levels of the bile acid tauro-β-muricholic acid in monocolonized gnotobiotic mice. B. thetaiotaomicron BSH mutant-colonized mice displayed altered metabolism, including reduced weight gain and respiratory exchange ratios, as well as transcriptional changes in metabolic, circadian rhythm, and immune pathways in the gut and liver. Our results demonstrate that metabolites generated by a single microbial gene and enzymatic activity can profoundly alter host metabolism and gene expression at local and organism-level scales.

Activity recognition is a fundamental research topic for a wide range of important applications such as fall detection for elderly people. Existing techniques mainly rely on wearable sensors, which may not be reliable and practical in real-world situations since people often forget to wear these sensors. For this reason, device-free activity recognition has gained the popularity in recent years. In this paper, we propose an RFID (radio frequency identification) based, device-free posture recognition system. More specifically, we analyze Received Signal Strength Indicator (RSSI) signal patterns from an RFID tag array, and systematically examine the impact of tag configuration on system performance. On top of selected optimal subset of tags, we study the challenges on posture recognition. Apart from exploring posture classification, we specially propose to infer posture transitions via Dirichlet Process Gaussian Mixture Model (DPGMM) based Hidden Markov Model (HMM), which effectively captures the nature of uncertainty caused by signal strength varieties during posture transitions. We run a pilot study to evaluate our system with 12 orientation-sensitive postures and a series of posture change sequences. We conduct extensive experiments in both lab and real-life home environments. The results demonstrate that our system achieves high accuracy in both environments, which holds the potential to support assisted living of elderly people.

Shenzhen, a coastal city, has changed from a small village to a supercity since the late 1980s. With the rapid development of its population and economy, ground disasters also occur frequently. These disasters bring great harm to human life and surface architecture. However, there is a lack of regular ground measurement data in this area. Permanent scatterer interferometry (PSI) technology can detect millimeter deformation of urban surface. In this paper, the building height and deformation from 2008 to 2010 in the Futian District of Shenzhen are obtained by using this technique alongside high-resolution TerraSAR-X data. For a visual expression of the result, we export the permanent scatterer (PS) points on buildings to Google Earth for 3D visualization after ortho-rectification of the PS height. Based on the Google Earth 3D model, the temporal and spatial characteristics of the building deformation became obvious. The InSAR measurements show that during the study period, the deformation rates of the Futian area are between −10 and 10 mm/year, and deformation is mainly distributed in a few buildings. These unstable activities can be attributed to human activities and the natural climate, which provides a reference for the local government to carry out a survey of surface deformation, as well as the monitoring and management of urban buildings, in the future.

Land use significantly influences the planet’s land surface and associated biogeochemical processes. With fierce conflict between various land uses, it is important to project the land system process to support decision-making. Lack of insight into scale differences of land use change (LUC) increased uncertainties in previous studies. To quantify the differences in LUCs within an elevation gradient, in this study, a novel model, the stratified land use change simulation model (SLUCS), was developed by using an elevation-based stratification strategy. This model consists of four modules. First, an elevation-based stratification module to develop a quantitative method for generating stratifications using elevation and land-use characteristics. Second, a non-spatial land-use demand module to forecast the overall land use area and make zoning constraints to simulate LUCs. Third, a stratified suitability estimation module that uses the stratified logistic regression method to reveal the regional relationship of the driving factors with LUCs at different stratifications. Finally, a spatial allocation of the land-use module, which projects a spatially explicit LUC. The SLUCS model was applied and tested in the Guizhou and Guangxi Karst Mountainous Region. Results validated the effectiveness of the model, and further demonstrated an improved spatial consistency with the reference, a higher accuracy assessment, and a better simulation performance in conversion areas than the traditional method. Three scenarios from 2015 to 2030 with different land-use priorities were designed and projected. Each scenario presented the same LUC trends, but with different magnitudes, including the rapid expansion of built-up land, the restoration of forest and water, and the loss of farmland and grassland. Priority of the socioeconomic development and ecological protection of the scenarios forecasted a sharper increase in the built-up land and in forests than the historical extrapolation scenario. The SLUCS model visually projected the LUC trajectory and competition between land uses, which suggests specific tradeoffs among management strategies to support sustainable land uses.

In this paper, a fault-tolerant control scheme is presented for a class of stochastic distribution collaborative control systems, which are composed of three subsystems connected in series to complete the control target. The radial basis function neural network is used to approximate the output probability density function of the third subsystem, which is also the output of the entire system. When fault occurs in the first subsystem, an adaptive diagnostic observer is designed to estimate the value of fault. However, the first subsystem does not have the ability of self-recovery, minimum rational entropy controllers are designed in the latter subsystems to compensate the influence of the fault and minimize the entropy of the system output. A numerical simulation is given to verify the effectiveness of the proposed scheme.

Falls are among the leading causes of hospitalization for the elderly and illness individuals. Considering that the elderly often live alone and receive only irregular visits, it is essential to develop such a system that can effectively detect a fall or abnormal activities. However, previous fall detection systems either require to wear sensors or are able to detect a fall but fail to provide fine-grained contextual information (e.g., what is the person doing before falling, falling directions). In this paper, we propose a device-free, fine-grained fall detection system based on pure passive UHF RFID tags, which not only is capable of sensing regular actions and fall events simultaneously, but also provide caregivers the contexts of fall orientations. We first augment the Angle-based Outlier Detection Method (ABOD) to classify normal actions (e.g., standing, sitting, lying and walking) and detect a fall event. Once a fall event is detected, we first segment a fix-length RSSI data stream generated by the fall and then utilize Dynamic Time Warping (DTW) based kNN to distinguish the falling direction. The experimental results demonstrate that our proposed approach can distinguish the living status before fall happening, as well as the fall orientations with a high accuracy. The experiments also show that our device-free, fine-grained fall detection system offers a good overall performance and has the potential to better support the assisted living of older people.