Search

Cancer is the most threat to human's health and life. At present, people have developed several ways to against cancer, such as surgical operation, radiotherapy and chemotherapy. However, cancers still cause highly mortality rate. A main part of the reason is that the traditional methods bring treatment effect as well as the negative effect. Recently, the anticancer peptides (ACPs) have been discovered which can be a new way to treat cancer. Since ACPs are natural biologics, they are safer than other methods. However, the experimental technology is an expensive way to find ACPs so we purpose a new machine learning method to identify the ACPs which named Random Relevance Vector Machines (RRVMs). The cross validations experiments show the high accuracy and stability of this new method.

Identification of new drug–target interactions (DTIs) is an important but a time-consuming and costly step in drug discovery. In recent years, to mitigate these drawbacks, researchers have sought to identify DTIs using computational approaches. However, most existing methods construct drug networks and target networks separately, and then predict novel DTIs based on known associations between the drugs and targets without accounting for associations between drug–protein pairs (DPPs). To incorporate the associations between DPPs into DTI modeling, we built a DPP network based on multiple drugs and proteins in which DPPs are the nodes and the associations between DPPs are the edges of the network. We then propose a novel learning-based framework, ‘graph convolutional network (GCN)-DTI’, for DTI identification. The model first uses a graph convolutional network to learn the features for each DPP. Second, using the feature representation as an input, it uses a deep neural network to predict the final label. The results of our analysis show that the proposed framework outperforms some state-of-the-art approaches by a large margin.

It is estimated that the impact of related genes on the risk of Alzheimer’s disease (AD) is nearly 70%. Identifying candidate causal genes can help treatment and diagnosis. The maturity of sequencing technology and the reduction of cost make genome-wide association study (GWAS) become an important means to find disease-related mutation sites. Because of linkage disequilibrium (LD), neither the gene regulated by SNP nor the specific SNP can be determined. Because GWAS is affected by sample size and interaction, we introduced empirical Bayes (EB) to make a meta-analysis of GWAS to greatly eliminate the bias caused by sample and the interaction of SNP. In addition, most SNPs are in the noncoding region, so it is not clear how they relate to phenotype. In this paper, expression quantitative trait locus (eQTL) studies and methylation quantitative trait locus (mQTL) studies are combined with GWAS to find the genes associated with Alzheimer disease in expression levels by pleiotropy. Summary data-based Mendelian randomization (SMR) is introduced to integrate GWAS and eQTL/mQTL data. Finally, we prioritized 274 significant SNPs, which belong to 20 genes by eQTL analysis and 379 significant SNPs, which belong to seven known genes by mQTL. Among them, 93 SNPs and 2 genes are overlapped. Finally, we did 10 case studies to prove the effectiveness of our method.

Genipin has been shown to exert anti-inflammatory effects, but its mechanism in protecting the ethanol-induced acute gastric injuries remains largely unclear. The present study aimed to investigate the effects of genipin on ethanol-induced acute gastric injuries in mice. After intragastrical administration of genipin for 7 consecutive days, acute gastric injuries were induced in the mice by ethanol treatment for 1 h. The expression levels of MDA, MPO, SOD, CAT, and NO in gastric tissues, and the levels of IL-6, TNF-α, MTL, SP, VIP and SS in serum samples were measured by ELISA. In addition, Western blotting was used to determine the expression levels of proteins involved in NLRP3 signaling pathway. The findings revealed that oral administration of genipin significantly ameliorated the pathological injury of gastric mucosa induced by ethanol, decreased the oxidative stress induced by ethanol and suppressed the expression levels of in-flammatory cytokines in gastric tissues and serum samples. In addition, it was observed that oral administration of genipin could remarkably inhibit the expression levels of related proteins in the NLRP3 signaling pathway. In conclusion, these results suggest that genipin may exhibit protective roles in ethanol-induced gastric mucosal injuries by activating antioxidant system and attenuating inflammatory reaction.

Engineering surface wettability is of great importance in academic research and practical applications. The exploration of hydrogel‐based natural surfaces with superior properties has revealed new design principles of surface superwettability. Gels are composed of a cross‐linked polymer network that traps numerous solvents through weak interactions. The natural fluidity of the trapped solvents confers the liquid‐like property to gel surfaces, making them significantly different from solid surfaces. Bioinspired gel surfaces have shown promising applications in diverse fields. This work aims to summarize the fundamental understanding and emerging applications of bioinspired gel surfaces with superwettability and special adhesion. First, several typical hydrogel‐based natural surfaces with superwettability and special adhesion are briefly introduced, followed by highlighting the unique properties and design principles of gel‐based surfaces. Then, the superwettability and emerging applications of bioinspired gel surfaces, including liquid/liquid separation, antiadhesion of organisms and solids, and fabrication of thin polymer films, are presented in detail. Finally, an outlook on the future development of these novel gel surfaces is also provided.

This paper proposed an optimization algorithm for torque sharing function (TSF) based on numerical iterative algorithm for switched reluctance motor (SRM). This optimization problem takes minimum rms voltage as the minimum optimization objective, current-limiting and bus voltage as the constraint. It gets free from a particular fitting function, making it has advantages of wide universality. Detailed design process of algorithm is illustrated in the paper. The results show that current-limiting is the main constraint at low speed. While at medium and high speeds, bus Voltage is the main limiting factor. The optimized TSF is applied to a 6/4 SRM. Simulations and experiment have been carried out to verify the effectiveness a of the proposed algorithm.

A fixed switching frequency current predict control method for switched reluctance motors (SRM) is presented in this paper. Utilizing the incremental inductance profile measured by the rotating method, the proposed current controller accurately predicts the required RMS voltage for a given reference current in next time step. Besides, considering the actual delay time of digital controller, this paper proposes a delay compensation method based on the iteration. Afterward, recursive least-squares (RLS) is used for resistance estimation while an adaptive controller is used to dynamically update CPC models. The control method is verified through simulation and followed by experimental validation based on TMS320F28335 DSP.

The effects of microplastics on aquatic organisms are drawing growing attention, but little has been focused on their effects on the toxicity of other chemicals. In this study, we examined the acute and chronic toxicity of micro-polystyrene (5.8 μm dia.), and its effects on the toxicity of three antimicrobial agents (triclosan, triclocarban and methyl-triclosan) to Daphnia magna. Results indicated that polystyrene had a low acute toxicity with an EC50 of 36.5 mg/L. The presence of polystyrene (1 mg/L) did not produce significant effect on the acute toxicity of three chemicals, because the 95% confidence intervals of their EC50 values had a large overlap of 11.3%–48.3%. For the 21 day chronic toxicity, polystyrene alone had significant toxicity with concentrations of at least 2 mg/L, which prolonged the time of the first brood, limited the number of broods, and reduced the total number of neonates. Compared with the chemicals alone, the addition of polystyrene enhanced their reproduction toxicity. Based on the various reproduction indicators, an intrinsic rate of natural increase (rm) was calculated to assess the rate of population growth. Results suggested that the rm values of three chemicals decreased in the presence of PS, and further decreased with increasing PS concentrations. Among the three chemicals, methyl-triclosan was the most affected. These results suggested that the presence of microplastics would exacerbate the detrimental influence of pollutants on Daphnia magna.

Background: More and more scholars are trying to use it as a specific biomarker for Alzheimer’s Disease (AD) and mild cognitive impairment (MCI). Multiple studies have indicated that miRNAs are associated with poor axonal growth and loss of synaptic structures, both of which are early events in AD. The overall loss of miRNA may be associated with aging, increasing the incidence of AD, and may also be involved in the disease through some specific molecular mechanisms. Objective: Identifying Alzheimer’s disease-related miRNA can help us find new drug targets, early diagnosis. Materials and Methods: We used genes as a bridge to connect AD and miRNAs. Firstly, proteinprotein interaction network is used to find more AD-related genes by known AD-related genes. Then, each miRNA’s correlation with these genes is obtained by miRNA-gene interaction. Finally, each miRNA could get a feature vector representing its correlation with AD. Unlike other studies, we do not generate negative samples randomly with using classification method to identify AD-related miRNAs. Here we use a semi-clustering method ‘one-class SVM’. AD-related miRNAs are considered as outliers and our aim is to identify the miRNAs that are similar to known AD-related miRNAs (outliers). Results and Conclusion: We identified 257 novel AD-related miRNAs and compare our method with SVM which is applied by generating negative samples. The AUC of our method is much higher than SVM and we did case studies to prove that our results are reliable.

To explore the evaluation method of evidence body for acupuncture and moxibustion clinical practice guidelines (CPGs), and to provide methodological support for the development of acupuncture and moxibustion CPGs and to promote the establishment and application of acupuncture and moxibustion standard. Based on the academic characteristics of TCM acupuncture, the stratified evidence evaluation method was proposed. Firstly, evidences were collected from the ancient literature, modern literature and clinical experience, and were scientifically distinguished and evaluated. Secondly, all the evidences were integrated as body of evidence by analytic hierarchy process (AHP). Lastly, based on the general index (GI), the clinical recommendations for acupuncture were proposed. Based on the stratified evidence evaluation method, acupuncture and moxibustion CPGs were developed in a more scientific and objectiveway with the comprehensive and sufficient evidences.

We propose a proactive dynamic network slicing scheme that utilizes a deep-learning based short-term traffic prediction approach for 5G transport networks. The demonstration shows utilization efficiency improvement from 46.33% to 71.53% under the evaluated scenario.

In this paper, a new steady-state power consumption model using the Elman Neural Network (ENN) is proposed. The model is dependent on the external parameters of chiller, which are easily monitored and which are related to the global optimization of an air-conditioning water system. The simulation results show that the model can complete the training process within 3 s. In addition, it can be seen that the results of the model are in good agreement with the experimental values with the majority of the RE values within ±3%. Therefore, this model is suitable for on-line prediction of the power consumption of chiller in on-field engineering.

In this paper, the online evaluation system for a chiller plant is proposed, which is significant for improving the energy efficiency and operation security of the chiller plant. For a general evaluation result, the system can estimate the chiller plant from three aspects: operating security, system performance, and cooling-supply quality, respectively. For each aspect, the paper presents a dynamic scoring method, which can convert the abstract evaluation indexes into the visualized contents by different evaluation demands. At the end of this paper, the effectiveness of the system is verified to perform excellently in practicability and operability, and it can also be used in most of the chiller plants.

Airway segmentation is a critical problem for lung disease analysis. However, building a complete airway tree is still a challenging problem because of the complex tree structure, and tracing the deep bronchi is not trivial in CT images because there are numerous small airways with various directions. In this paper, we develop two-stage 2D+3D neural networks and a linear programming based tracking algorithm for airway segmentation. Furthermore, we propose a bronchus classification algorithm based on the segmentation results. Our algorithm is evaluated on a dataset collected from 4 resources. We achieved the dice coefficient of 0.94 and F1 score of 0.86 by a centerline based evaluation metric, compared to the ground-truth manually labeled by our radiologists.

Correlation models between human behavior and energy consumption of teaching buildings are crucial for devising energy efficiency strategies. Especially in teaching buildings in campus, on one hand, the investment is limited and the level of operation and maintenance is low. On the other hand, the certain management system exists, and the flow of occupant in campus is also regular obviously. So, it is difficult for large-scale technical retrofits and possible for management retrofits. In order to the low-cost management retrofits, the model between human behavior and energy consumption are studied in this paper. According to the models, on the one hand, managers of campus can predict energy consumption based in historical, for matching energy consumption supply and demand each other. On the other hand, managers can judge future human behaviors based on historical energy consumption, for conducting management to reduce energy consumption. In this paper, first, correlation models between the energy consumption and human behavior are proposed and the lighting management level coefficient K is defined to evaluate the lighting management level. With the aid of the correlation models and K, managers can obtain management strategies to increase energy efficiency without large-scale technical retrofits. The verification results show that the proposed strategies could bring a 13.96% economic benefit ratio. Second, correlation models between human behavior and influencing factors of human behavior are proposed for predicting human behavior. Finally, a technical route of energy efficiency retrofit by management is suggested and validated. Although the energy-saving ratio is not conspicuous, the cost is low and the operation is easy. It means that the technical route is worthy for research and promotion.

Compared to traditional hydrogel materials, organohydrogels have wider applications due to their special properties such as abilities of self-adaptive, freeze-tolerant and mechanically stable over a wide range temperature. In this paper, we prepared Poly (vinyl alcohol)/Poly (butyl methacrylate)-co-Poly (lauryl methacrylate) multiple network organohydrogels (MN-OHGs) by multiple swelling polymerization method. By tuning the components of heterogeneous networks, the swelling behaviors and the surface wettability in different solvents of MN-OHGs could be controlled. When the ratio of oleophilic polymer network (OPN) content to hydrophilic polymer network (HPN) content was 1:1.4, the MN-OHGs showed the highest mechanical properties (the elastic modulus could reach 8 MPa, and the toughness could reach 12 MJ/m3). Moreover, when the OPN/HPN ratio was 2/1, the MN-OHGs could maintain a stable volume in artificial seawater with mechanical properties enhanced by 47.3%. These characteristics demonstrated that the MN-OHGs were ideal candidates for functional materials which could be applied in complex environments.

Semantic Web has recently gained traction with the use of Linked Open Data (LOD) on the Web. Although numerous state-of-the-art methodologies, standards, and technologies are applicable to the LOD cloud, many issues persist. Because the LOD cloud is based on graph-based resource description framework (RDF) triples and the SPARQL query language, we cannot directly adopt traditional techniques employed for database management systems or distributed computing systems. This paper addresses how the LOD cloud can be efficiently organized, retrieved, and evaluated. We propose a novel hybrid approach that combines the index and live exploration approaches for improved LOD join query performance. Using a two-step index structure combining a disk-based 3D R*-tree with the extended multidimensional histogram and flash memory-based k-d trees, we can efficiently discover interlinked data distributed across multiple resources. Because this method rapidly prunes numerous false hits, the performance of join query processing is remarkably improved. We also propose a hot-cold segment identification algorithm to identify regions of high interest. The proposed method is compared with existing popular methods on real RDF datasets. Results indicate that our method outperforms the existing methods because it can quickly obtain target results by reducing unnecessary data scanning and reduce the amount of main memory required to load filtering results.

The establishment of an accurate energy consumption prediction model is an important way to analyze the energy consumption of a campus building, thereby providing a reliable basis for the selection of energy-saving transformation measures. To establish a high-precision campus energy consumption model, this paper proposes an improved bottom-up energy consumption prediction method based on energy consumption monitoring data and the Bayesian theory. The measured data of the monitoring platform are used as sample information in the Bayesian analysis. Considering a campus complex in China as an example, a traditional model using the area expansion method and the improved model has been established in this paper. For a certain type of building, the prediction results indicate that the maximum monthly error rate of the traditional model is 54% while that of the improved model is within 15%. This indicates that the improved model has advantages with regard to prediction accuracy. In addition, considering both the construction cost of the energy consumption monitoring platform and the prediction accuracy, this study explores the optimal sample number in the Bayesian analysis. Based on the analysis of different prior information and different research objects, with an error of 20% as the standard, the optimal sample size was finally determined to be four.

An effective indoor temperature model would assist in improving energy efficiency and indoor thermal comfort of air conditioning system. However, it is difficult to build an accurate model due to lag response characteristic in the regulation process of indoor temperature. To solve this problem, the modeling and prediction methods for indoor temperature lag response characteristic based on time-delay neural network (TDNN) and Elman network neural (ENN) are presented. Then, taking variable air volume (VAV) air conditioning system as the study object, the effectiveness and practicability of proposed methods are validated using simulation sampling data and real-time operating data. Results indicate that ENN could be considered as a better modeling method for indoor temperature prediction for its simpler network structure, smaller storing space and better prediction accuracy. The contribution of this study is to provide an applicable online ANN modeling method for indoor temperature lag characteristic, and detailed training and validation for online implementation are presented, which will benefit for engineers and technicians to use in practical engineering. Meanwhile, this study provides the reference for online application of advanced intelligent algorithms in the building engineering.

Background Puffball is a common Chinese medicine of edible and medicinal fungus. It promotes hemostasis and muscle regeneration. Recent studies have shown that puffball has a limiting effect on pain and inflammation, but the specific mechanism behind this action remains unclear. This study aimed to reveal the pharmacological mechanism of the anti-inflammatory effect of puffball using the network pharmacology method. Methods TCMSP and SwissTargetPrediction were used to analyze and predict the active ingredients and corresponding targets in puffball. The DisGeNET and GeneCards databases were searched to obtain inflammation-related targets, and the PPI network of the cross-genes of puffball ingredients and diseases was constructed through STRING and Cytoscape. The Metascape database was used for GO function enrichment analysis of co-presence genes and KEGG signal pathway enrichment analysis to construct an “ingredient-target-pathway” network and analyze its mechanism of action. Meanwhile, molecular docking was used to verify the ability of active ingredients to bind to key targets. Results A total of 6 inflammation-related active ingredients were screened from the 32 candidate compounds of puffball, and 236 targets that correspond to ingredients and 328 inflammation-related targets were collected. In summary, the β-sitosterol and melanin components in puffball can act on signal transducers and activators of transcription (STAT3) and epidermal growth factor receptor (EGFR) targets and are mainly differentiated through Th17 cells signal pathway and cancer-related pathway to play an anti-inflammatory effect. Conclusion This study has successfully predicted the active ingredients and potential targets of Puffball's anti-inflammatory effects, which helped to systematically elucidate its mechanism of action. The network pharmacology method can provide new insights for the research and development of edible and medicinal fungi.

The automatic control of the existing indoor thermal environment is usually based on the user's set points without considering their real-time thermal feeling, causing an uncomfortable thermal environment and high energy consumption. In this study, the wristband device's physiological data is utilized to propose a thermal sensation prediction-based fuzzy control method of indoor temperature to improve the indoor environment's thermal comfort and achieve an energy saving of air-conditioning system. The linear regression model-based updating method is utilized to update the thermal sensation prediction model in real-time. Moreover, Mamdani fuzzy model and FFSI method are adopted to develop the fuzzy control algorithm for indoor temperature set point optimization. Several comparative experiments are performed to compare control performance, thermal comfort, and energy consumption of thermal sensation prediction-based control with thermal sensation feedback-based control and temperature set point-based control. The results show that thermal sensation prediction-based control can adjust the temperature set point by monitoring physiological data from the subjects without interfering with their regular works. It is also revealed that the thermal comfort performance of thermal sensation prediction-based control is superior to the temperature set point-based control, while it is similar to the thermal sensation feedback-based control. Furthermore, the proposed control approach provides 20.07% and 10.73% savings in daily energy consumption compared with the temperature set point-based control and thermal sensation feedback-based control, respectively.

This paper aims to analyze the effects of mineralogy on petrophysical properties and evaluate the performance of eight permeability estimation models in tight oil sandstones. The results include mineralogy, pore size distribution, porosity, and permeability. The evaluated eight permeability models are Purcell, Winalnd, Swanson, Thomeer, Wells-Amaefule (W-A), Katz-Thompson (K-T), Pittman and Dastidar models. The experimental results indicate that the samples are rich in quartz (averages 41.5%), feldspar (30.0%), and clays (18.5%); have a wide pore size distribution mainly in the nanometer scale (9.2 nm–1 μm); and have low porosity (11.07% on average) and ultralow permeability (0.22 mD). Quartz and feldspar preserve the primary pores by supporting the overburden pressure, and feldspar further enhances the porosity by forming secondary dissolution pores. Mixed layer of illite/smectite (I/S) and illite significantly decrease porosity while chlorite preserves the primary pores by inhibiting silica overgrowth. The Pittman model has the best estimation accuracy ( R adj 2 = 0.968); the Winland, K-T and Dastidar models give reasonable estimations ( R adj 2 > 0.92); others have poor performance ( R adj 2

This paper discusses a project selection and adjustment problem in the situation where some project parameters are given by experts’ estimates because of lack of historical data. Uncertain variables are used to describe these project parameters and the use of them is justified. Based on uncertain measure, a cost overrun risk which describes the average amount of investment outlay exceeding the available budget is proposed, and a new optimization model which simultaneously considers the selection of new projects and the adjustment of existing ones is developed. To solve the proposed problem, the deterministic equivalents of the model are provided and a genetic algorithm is offered. As an illustration, an example is also presented and discussed.

Accurate micro-storage and mechanism of shale gas is important for defining and describing the adsorption characteristics, and for gas-in-place estimation. In this paper, reservoir forming process is divided into early stage and late stage, and the initial gas-water distribution in different stage is redefined. Different from the widely used solid-gas interface adsorption theory, the pore system in original shale reservoir is divided into solid-gas system and solid-liquid-gas system. Gas pores existing in organic matters belong to solid-gas system, and are classified into residual gas pores and metamorphic gas pores. Plant tissue pores in organic and mineral pores belong to solid-liquid-gas system. Their characteristics are described.

Shale gas has attracted increasing attention as a new potential source of natural gas in recent years. Accurate micro-storage and mechanism of shale gas is important for defining and describing the adsorption characteristics, gas-in-place estimates. Organic-rich shale gas reservoirs were usually deposited under water environment which has great impact on shale gas adsorption. Different from the widely used solid-gas interface adsorption theory, the pore system in original shale reservoir is divided into solid-gas system and solid-liquid-gas system. In same statement, shale pores have stronger interaction with water molecules than CH4 molecules, especially in solid-gas system. Water molecules mainly formed as competitive adsorption with shale gas on solid surface (solid-gas system) and liquid water surface (solid-liquid-gas system). Maxim gravitational potential energy, Ψ max, the distance where Ψ max occurs, z m and limited pore side, D L are given. When pore size D < D L, no matter in what position the adsorbed molecules is, it force by both sides of the pore silts. When D > D L, the ratio of unaffected area increases to approximately 100% with pore side. When D < 10 nm, the ratio increases rapidly this means that the ratio of free gas increases quickly. When D > 50 nm, the ratio of unaffected area almost reaches 100%, which means that the content of free gas is far greater than the adsorbed gas. This study not only provides a new classification of shale pore system, but also can act as a reference to better understanding the adsorption mechanism of shale reservoir.

Sensor errors significantly affect the operation and control of building energy systems. Thus, using correct and reliable sensors can effectively reduce the energy consumption of building energy systems. Virtual in-situ calibration (VIC), which is based on Bayesian inference and the Markov Chain Monte Carlo method, can help avoid the installation of new sensors, effectively reduce the systematic and random errors of sensors, and improve the reliability of collected data. In the current study, six working conditions were designed to check the robustness and accuracy of the proposed VIC technology. It's shown that for the systematic and random error of various sensors, the improved component calibration method is the most accurate. Contrarily, the whole calibration method is the least accurate, whereas the performance of the three local calibrations is intermediate. The maximum errors of the whole, local, and component calibration methods were 973%, 112.7%, and 30%, respectively. This verifies that component calibration effectively reduces the possibility of abnormal data and also enhances the reliability of sensor measurements. For the whole and local calibration methods, the constraints between various parameters are significantly reduced because of the random and uncertain coefficients; consequently, a majority of the correction results deviate from their true values. It is not necessary to determine the sensitivity and normalizing coefficients through a complicated optimization algorithm or historical experience by using the proposed component calibration. Therefore, the component calibration method features advantages in terms of computational time and correction accuracy.

Brucellosis is a zoonotic infectious disease caused by Brucella infection. MarR-family transcription factors are closely related to diverse physiological functions necessary for many pathogens adaptation to environmental changes. However, whether the MarR-family transcription factors are involved in virulence, mediated inflammatory responses and regulated virulence gene expression in the intracellular pathogen Brucella are still unknown. Therefore, we created a 2308ΔMarR6 mutant of B. abortus 2308 (S2308). Virulence and inflammatory cytokines assays were performed using a murine macrophage cell line (RAW 264.7). We also performed chromatin immunoprecipitation of MarR6 followed by next-generation sequencing (ChIP-seq). The results showed that 2308ΔMarR6 was significantly reduced survival capability in RAW 264.7. After the macrophages were infected with 2308ΔMarR6, the levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6), interleukin-12 (IL-12), interferon-gamma (IFN-γ) and macrophage chemoattractant protein-1 (MCP-1) were decreased and were significantly lower than that for the S2308-infected group, indicating that the 2308ΔMarR6 mutant could reduce the secretion of inflammatory cytokines. Furthermore, we detected 122 intergenic ChIP-seq peaks of MarR6 binding distributed across the Brucella genome. Taken together, the research has recorded valuable data about MarR6. Our findings are of great significance in elucidating the function of MarR6.

Accurate and reliable measurements collected from various physical sensors are of great importance to improve the energy utilization efficiency of solar energy systems. Sensor errors have various significant negative impacts on the automatic control and optimized operation. The current study used the virtual in-situ calibration (VIC) instead of traditional methods to spot possible systematic and random errors of temperature, pressure and mass flowrate sensors in a photovoltaic-thermal heat pump (PVT-HP) system. Seven working conditions were designed to check the robustness and accuracy of the proposed VIC technology. The effects of systematic and random errors on the sensor calibration process and performance estimation of a PVT-HP system were investigated. It’s shown that after calibration the systematic errors of all kinds of sensors were less than 2% and the random errors were also reduced by as much as 74%, significantly reducing the possibility of abnormal data and enhancing the reliability of sensor measurements. The VIC method is very effective to guarantee the thermal performance of PVT-HP system.

Hydrogels have drawn great attentions in the past two decades due to their excellent biocompatibility and multi stimuli responsiveness, which have a wide range of applications in the field related to tissue engineering, sensor and biomedicine. However, conventional artificial hydrogels are usually isotropic in structure with random crosslinking of polymer chains. To imitate the well-defined hierarchical structures ranging from the molecular scale to macroscopic scale like biological soft tissues in hydrogels. Herein, an anisotropic thermoresponsive hydrogel was reported via a linear remolding of highly stretchable clay-PNIPAm nanocomposite hydrogel by a secondary crosslinking. The as-prepared hydrogel shows anisotropic mechanical performance and can deformed anisotropically in response to temperature change. Besides, the oriented structures of clay nanosheets and polymer network that contribute to understand the anisotropic mechanism of SC-hydrogel was investigated. The special functions of current SC-hydrogel suggest that it may serve as ideal composite gel materials with validity in a variety of applications, such as soft actuators, sensors, and biological materials.

In nature, many adaptive organohydrogels with synergistic heterostructures exist extensively in biological soft tissues and play an important role in biological activities, such as anti-freezing organohydrogels in alpine organism cells and mechanical anisotropic organohydrogels in muscle tissues. The extraordinary adaptive function of natural hydrogels far exceeds that of current synthetic hydrogels. Therefore, bioinspired design is an effective approach to develop high-performance hydrogels with adaptability. Until now, several empirical strategies have developed to fabricate the adaptive hydrogels. In this feature article, we reviewed our recent advances in high-performance adaptive organohydrogels with synergistic heterostructures in terms of critical design principles as well as fabrication strategies, including freezing tolerance, surface functionalization, shape memory, and mechanical anisotropy. Meanwhile, we also explored the current challenges and prospects of the bioinspired adaptive hydrogel field.

In this paper, we consider supply chain coordination with a revenue sharing contract (RSC) and a linear quantity discount contract (LQDC) in a fashion supply chain with demand disruptions. The situation without demand disruptions is considered as the benchmark. In the benchmark, we find that, the RSC fails to coordinate the supply chain while the LQDC does. We also find that demand disruptions can promote supply chain coordination. We then find that, in some cases, the RSC (LQDC) fails to coordinate the supply chain whereas the LQDC (RSC) can do so.

In multi-zone variable outdoor air (OA) volume air-conditioning systems, there is often a mismatch between OA volume flow supply and demand, because variable air volume (VAV) systems exhibit nonlinear, multivariable, and long delay characteristics. Traditional methods ensure indoor air quality (IAQ) by increasing OA volume flow, which leads to a waste of OA flow energy. To address this problem, an optimal control method for determining discrete variable OA volume flow setpoints, based on differential pressure control, is proposed in this study. In this method, changes in indoor CO2 concentration and the total air volume flow demand of terminals are regarded as constraints of OA volume flow setpoints. A Proportional-Integral (PI) calculation method and trial-and-error method are used to discretize continuous OA volume flow setpoints to increase system stability. Two types of experiments were conducted to compare the characteristics of the two discrete OA volume flow setpoint determining methods, and to verify the feasibility of our new approach. The results show that the OA volume flow control loop can be well fitted with other control loops of VAV systems, and ensure OA volume flow demand for a target area. The proposed method provides greater energy savings than maximum OA ratio control strategy (7%–22%) and typical on-demand ventilation method (6.1%), and better control performance than continuous OA volume flow setpoints control system. The PI calculation method is suitable for conditions with high IAQ requirements, while the trial-and-error method can be used for comfort air-conditioning systems to adjust the indoor environment.

Person re-identification is an important technique towards automatic search of a person’s presence in a surveillance video. It is becoming a hot research topic due to its value in both machine learning research and video surveillance applications. Considering the current success of deep learning, having tons of person images with identity labels are important and helpful for learning effective person matchers. However, collecting labeled images for person re-identification is more difficult than other similar tasks such as face recognition due to complex intra-class variations in illumination, pose, viewpoint, blur, low resolution, and occlusion. Although the volume of surveillance videos has become larger and larger today, it is time-consuming and costs lots of human labors in labeling a large dataset for person re-identification. In this paper, we propose a semi-automatic data annotation tool to accelerate annotation of person images across multi cameras. This tool consists of automatic person detection and tracking algorithms for person image collection, and an ad-hoc person matcher for automatic person matching suggestions across multi cameras. Moreover, we further utilize background and video sequence information for identity confirmation during annotation, which is also a good intuition for the future design of person re-identification algorithms.

Alzheimer’s disease (AD) is a type of neurodegenerative disease that has become the fourth “health killer” for the elderly after cardiovascular, malignant, and stroke. Identifying AD-related proteins can effectively help diagnose diseases in advance, discover new drug targets, and gain a deeper understanding of the pathogenesis of disease. Due to the high cost of biological experiments, more and more researchers introduced advanced algorithms into this field. Based on the hypothesis of “similar diseases shares similar related proteins in this paper, diseases which are similar to AD were found by five similarity calculation methods. Then, related proteins of each disease were obtained by public data set. Through these proteins, features are extracted. Then, these features are mapped to disease similarity by Logistic Regression (LR). To avoid curse of dimensionality, random features are selected each time and hundreds of models were built. For each model, AD-related proteins could be obtained by Gradient Descent method. Finally, we integrated all the models and get all the proteins related to AD by weight counting. Finally, we did three case studies to prove novel proteins found by us are reliable.

Alzheimer disease (AD) is the fourth major cause of death in the elderly following cancer, heart disease and cerebrovascular disease. Finding candidate causal genes can help in the design of Gene targeted drugs and effectively reduce the risk of the disease. Complex diseases such as AD are usually caused by multiple genes. The Genome-wide association study (GWAS), has identified the potential genetic variants for most diseases. However, because of linkage disequilibrium (LD), it is difficult to identify the causative mutations that directly cause diseases. In this study, we combined expression quantitative trait locus (eQTL) studies with the GWAS, to comprehensively define the genes that cause Alzheimer disease. The method used was the Summary Mendelian randomization (SMR), which is a novel method to integrate summarized data. Two GWAS studies and five eQTL studies were referenced in this paper. We found several candidate SNPs that have a strong relationship with AD. Most of these SNPs overlap in different data sets, providing relatively strong reliability. We also explain the function of the novel AD-related genes we have discovered.

To improve the dynamic performance of the current loop, a robust deadbeat predictive control (RPC) scheme for dual three-phase permanent magnet synchronous motor (DTP-PMSM) is presented. Based on the mathematical model of DTP-PMSM, the current controller directly calculates the expected voltage vector according to the current reference and feedback values, then transformed the voltage vector into switching signals by four-dimensional three-phase decomposition space vector pulse width modulation (SVPWM). The effectiveness and validity of the robust deadbeat predictive controller are verified through experiment on a 750W DTP-PMSM.

This paper proposes a SDN-based cross-layer network slicing architecture for service migration in the mobile edge computing infrastructure. Experimental demonstration shows that the scheme realizes cross-layer resource slicing and service migration isolation capabilities.

We proposed a flexible architecture and scheduling scheme across TDM/WDM/SDM dimensions for access network sharing architecture to support network slicing. The demonstration shows that the time and space dimensional settling time is 114ms and 177ms.

In this paper, a deep mixture of diverse experts algorithm is developed to achieve more efficient learning of a huge (mixture) network for large-scale visual recognition application. First, a two-layer ontology is constructed to assign large numbers of atomic object classes into a set of task groups according to the similarities of their learning complexities, where certain degrees of inter-group task overlapping are allowed to enable sufficient inter-group message passing. Second, one particular base deep CNNs with $M+1$ M + 1 outputs is learned for each task group to recognize its $M$ M atomic object classes and identify one special class of “not-in-group”, where the network structure (numbers of layers and units in each layer) of the well-designed deep CNNs (such as AlexNet, VGG, GoogleNet, ResNet) is directly used to configure such base deep CNNs. For enhancing the separability of the atomic object classes in the same task group, two approaches are developed to learn more discriminative base deep CNNs: (a) our deep multi-task learning algorithm that can effectively exploit the inter-class visual similarities; (b) our two-layer network cascade approach that can improve the accuracy rates for the hard object classes at certain degrees while effectively maintaining the high accuracy rates for the easy ones. Finally, all these complementary base deep CNNs with diverse but overlapped outputs are seamlessly combined to generate a mixture network with larger outputs for recognizing tens of thousands of atomic object classes. Our experimental results have demonstrated that our deep mixture of diverse experts algorithm can achieve very competitive results on large-scale visual recognition.

In this paper, a level-wise mixture model (LMM) is developed by embedding visual hierarchy with deep networks to support large-scale visual recognition (i.e., recognizing thousands or even tens of thousands of object classes), and a Bayesian approach is used to adapt a pre-trained visual hierarchy automatically to the improvements of deep features (that are used for image and object class representation) when more representative deep networks are learned along the time. Our LMM model can provide an end-to-end approach for jointly learning: (a) the deep networks to extract more discriminative deep features for image and object class representation; (b) the tree classifier for recognizing large numbers of object classes hierarchically; and (c) the visual hierarchy adaptation for achieving more accurate indexing of large numbers of object classes hierarchically. By supporting joint learning of the tree classifier, the deep networks and the visual hierarchy adaptation, our LMM algorithm can provide an effective approach for controlling inter-level error propagation effectively, thus it can achieve better accuracy rates on large-scale visual recognition. Our experiments are carried on ImageNet1K and ImageNet10K image sets, and our LMM algorithm can achieve very competitive results on both the accuracy rates and the computation efficiency as compared with the baseline methods.

Knowledge of pore structure and adsorption capacity provides guidance for better studying the origin, hydrocarbon distribution, and productivity of shale gas reservoir. In this study, pore structure characteristics of six shale core plugs with different maturity from the Lower Silurian Longmaxi formation in south China were investigated using the Rock-eval analysis, X-ray diffraction, total organic carbon (TOC) content test, and scanning electron microscope (SEM) observation. To further investigate the influence of maturity, the adsorption behavior of gas shale was experimentally measured, with the maximal pressure being 20 MPa. Rock-eval analysis indicates that Ro is 0.67~1.34%. SEM results show that organic matter (OM) pores are abundant in high-maturity shale sample. The OM pores are mainly irregular to elliptical in shape, the size is 8~100 nm. The TOC content is 0.16~4.21% and shows a positive correlation with the BET surface area. A negative relationship exists between TOC content and average pore diameter, which indicates that abundant nanometer pores are related to the OM. A noticeable characteristic in the pore size distribution curve is that the content of micropores (pore width

Lung segmentation is usually the first step of lung CT image analysis and plays an important role in lung disease diagnosis. We propose an efficient end-to-end fully convolutional neural network to segment lungs with different diseases in CT images. We introduce a multi-instance loss and a conditional adversary loss to the neural network in order to solve the segmentation problem for more severe pathological conditions. Our method is capable of solving the lung segmentation problem under normal, moderate and severe pathological conditions, which is validated on 3 public benchmark datasets with different diseases.

Conventional synthetic hydrogels are intrinsically soft and brittle, which severely limits the scope of their applications. A variety of approaches have been proposed to improve the mechanical strength of hydrogels. However, a facile and ubiquitous strategy to prepare hydrogels with high mechanical strength and stiffness is still a challenge. Here, we report a general strategy to prepare highly filled microcomposite hydrogels with high mechanical performance using an ultrasonic assisted strategy. The microparticles were dispersed in the polymer network evenly, resulting in homogeneous and closely packed structures. The as-prepared hydrogels with extraordinary mechanical performance can endure compressive stress up to 20 MPa (at 75% strain) and exhibit high stiffness (elastic modulus is around 18 MPa). By using our comprehensive strategy, different hydrogels can enhance their mechanical strength and stiffness by doping various microparticles, leading to a much wider variety of applications.

Background More than 1/3 of human genes are regulated by microRNAs. The identification of microRNA (miRNA) is the precondition of discovering the regulatory mechanism of miRNA and developing the cure for genetic diseases. The traditional identification method is biological experiment, but it has the defects of long period, high cost, and missing the miRNAs that but also many other algorithms only exist in a specific period or low expression level. Therefore, to overcome these defects, machine learning method is applied to identify miRNAs. Results In this study, for identifying real and pseudo miRNAs and classifying different species, we extracted 98 dimensional features based on the primary and secondary structure, then we proposed the BP-Adaboost method to figure out the overfitting phenomenon of BP neural network by constructing multiple BP neural network classifiers and distributed weights to these classifiers. The novel method we proposed, from the 4 evaluation terms, have achieved greatly improvement on the effect of identifying true pre-RNA compared to other methods. And from the respect of identifying species of pre-RNA, the novel method achieved more accuracy than other algorithms. Conclusions The BP-Adaboost method has achieved more than 98% accuracy in identifying real and pseudo miRNAs. It is much higher than not only BP but also many other algorithms. In the second experiment, restricted by the data, the algorithm could not get high accuracy in identifying 7 species, but also better than other algorithms.

The low contrast and irregular cell shapes in microscopy images cause difficulties to obtain the accurate cell segmentation. We propose pyramid-based fully convolutional networks (FCN) to segment cells in a cascaded refinement manner. The higher-level FCNs generate coarse cell segmentation masks, attacking the challenge of low contrast between cell inner regions and the background. The lower-level FCNs generate segmentation masks focusing more on cell details, attacking the challenge of irregular cell shapes. The FCNs in the pyramid are trained in a cascaded way such that the residual error between the ground truth and upper-level segmentation is propagated to the lower-level and draws the attention of the lower-level FCNs to find the cell details missed from the upper-levels. The fine cell details from lower-level FCNs are gradually fused into the coarse segmentation from upper-level FCNs so as to obtain a final precise cell segmentation mask. On the ISBI cell segmentation challenge dataset and a newly collected dataset with high-quality ground truth, our method outperforms the state-of-the-art methods.

We proposed a multi-tenant slicing scheme with multi-domain, multi-layer and multivendor (M3) features. The experimental demonstration shows the slicing created and modified around in 353ms and 150ms, with the hybrid resource isolation has offered.

In variable air volume (VAV) systems, conventional control methods reset the supply air static pressure based on the exact mathematics model of the controlled objects which are often difficult to be founded, because the VAV controlled systems have the characteristics of nonlinear, multivariable and long delay times. In order to develop a simple inference process for real-time applications, this paper presents a novel fuzzy control method for static pressure reset with no using of the mathematics model. The proposed fuzzy control method is developed based on the maximum damper position feedback using Mamdani-type fuzzy rule and functioning fuzzy subset inference (FFSI). This study provides the detail implement process of the proposed fuzzy control method. Supported by the experimental data, the proposed method is compared with optimal fixed changing step method in an integrated control test rig for a VAV air conditioning system. Experimental results show that the proposed method has better energy efficient and control performance. A minimum of 7% energy saving of the proposed fuzzy control method is shown through experimental study.