Design, optimization, and crosstalk analysis for robot joint torque sensor of shear beam structure.

Crosstalk resistance is an important criterion for evaluating the measurement error of the Joint Torque Sensor (JTS) in actual collaborative robot application, but few research literature studies on the crosstalk resistance of shear beam-type JTS have been found. This paper proposes a mechanical structure of one shear beam sensor and determines its strain gauge working area. Multi-objective optimization equations are established with three major performance indicators of sensitivity, stiffness, and crosstalk resistance. Optimal processing and manufacturing structure parameters are obtained by employing both the response surface method based on the central composite design experimental principle and the multi-objective genetic algorithm. By simulating and experimenting, the optimized sensor is verified and has the following indices: overload resistance 300% F.S., torsional stiffness 503.44 KN m/rad, bending stiffness 142.56 KN m/rad, range 0-±200 N m, sensitivity 25.71 mV/N m, linearity 0.1999%, repeatability error 0.062%, hysteresis error 0.493%, measurement error less than 0.5% F.S. under Fx (392.4 N) or Fz (600 N) crosstalk load, and measurement error less than 1% F.S. under My (25 N m) moment crosstalk. The proposed sensor possesses good crosstalk resistance and especially axial crosstalk resistance and has good overall performance to meet well the engineering requirements.

Identification of non-coding RNA related prognosis biomarkers based on ceRNA network in thyroid cancer.

Introduction: Thyroid cancer (THCA) has become a serious malignant tumor worldwide. Identification of non-coding RNA related regulators is very necessary to improve the knowledge of THCA treatment. The aim of this study was to identify novel therapeutic targets and prognosis biomarkers for predicting pathological characteristics and subsequently treating THCA. Methods: We investigated the alterations of miRNAs, mRNAs and lncRNAs in THCA. Functional enrichment and clustering analysis were conducted for these aberrantly expressed RNAs. Multiple interaction networks among miRNAs, mRNAs and lncRNAs were constructed and the functional modules associated with THCA patients' prognosis were identified. Furthermore, we evaluated the prognostic roles of the important miRNAs, mRNAs and lncRNAs in THCA and investigated the regulatory potential of non-coding RNAs on immune cell infiltration. Results: We firstly identified that miR-4709-3p and miR-146b-3p could significantly classify patients into high/low risk groups, which may be potential prognosis biomarkers of THCA. Secondly, we constructed a THCA-related miRNA-mRNA network, which displayed small world network topological characters. Two THCA-related functional modules were identified from the miRNA-mRNA network by MCODE. Results showed that two modules could implicate in known cancer pathways, such as apoptosis and focal adhesion. Thirdly, a THCA-related miRNA-lncRNA network was constructed. A subnetwork of miRNA-lncRNA network showed strong prognosis effect in THCA. Fourthly, we constructed a THCA-related mRNA-lncRNA network and detected several typical lncRNA-miRNA-mRNA crosstalk, such as AC068138, BCL2, miR-21 and miR-146b, which had good prognosis effect in THCA. Immune infiltration results showed that lncRNAs LA16c-329F2, RP11-395N3, RP11-423H2, RP11-399B17 and RP11-1036E20 were high related to neutrophil and dendritic cell infiltration. Discussion: Non-coding RNA-mediated gene regulatory network has the strong regulatory potential in pathological processes of THCA. All these results could help us uncover the non-coding RNA-mediated regulatory mechanism in THCA.

Few-shot remote sensing image scene classification based on multiscale covariance metric network (MCMNet).

Few-shot learning (FSL) is a paradigm that simulates the fast learning ability of human beings, which can learn the feature differences between two groups of small-scale samples with common label space, and the label space of the training set and the test set is not repeated. By this way, it can quickly identify the categories of the unseen image in the test set. This method is widely used in image scene recognition, and it is expected to overcome difficulties of scarce annotated samples in remote sensing (RS). However, among most existing FSL methods, images were embed into Euclidean space, and the similarity between features at the last layer of deep network were measured by Euclidean distance. It is difficult to measure the inter-class similarity and intra-class difference of RS images. In this paper, we propose a multi-scale covariance network (MCMNet) for the application of remote sensing scene classification (RSSC). Taking Conv64F as the backbone, we mapped the features of the 1, 2, and 4 layers of the network to the manifold space by constructing a regional covariance matrix to form a covariance network with different scales. For each layer of features, we introduce the center in manifold space as a prototype for different categories of features. We simultaneously measure the similarity of three prototypes on the manifold space with different scales to form three loss functions and optimize the whole network by episodic training strategy. We conducted comparative experiments on three public datasets. The results show that the classification accuracy (CA) of our proposed method is from 1.35 % to 2.36% higher than that of the most excellent method, which demonstrates that the performance of MCMNet outperforms other methods.

Curriculum Reinforcement Learning Based on K-Fold Cross Validation.

With the continuous development of deep reinforcement learning in intelligent control, combining automatic curriculum learning and deep reinforcement learning can improve the training performance and efficiency of algorithms from easy to difficult. Most existing automatic curriculum learning algorithms perform curriculum ranking through expert experience and a single network, which has the problems of difficult curriculum task ranking and slow convergence speed. In this paper, we propose a curriculum reinforcement learning method based on K-Fold Cross Validation that can estimate the relativity score of task curriculum difficulty. Drawing lessons from the human concept of curriculum learning from easy to difficult, this method divides automatic curriculum learning into a curriculum difficulty assessment stage and a curriculum sorting stage. Through parallel training of the teacher model and cross-evaluation of task sample difficulty, the method can better sequence curriculum learning tasks. Finally, simulation comparison experiments were carried out in two types of multi-agent experimental environments. The experimental results show that the automatic curriculum learning method based on K-Fold cross-validation can improve the training speed of the MADDPG algorithm, and at the same time has a certain generality for multi-agent deep reinforcement learning algorithm based on the replay buffer mechanism.

Fast Task Adaptation Based on the Combination of Model-Based and Gradient-Based Meta Learning.

Deep reinforcement learning (DRL) recently has attained remarkable results in various domains, including games, robotics, and recommender system. Nevertheless, an urgent problem in the practical application of DRL is fast adaptation. To this end, this article proposes a new and versatile metalearning approach called fast task adaptation via metalearning (FTAML), which leverages the strengths of the model-based methods and gradient-based metalearning methods for training the initial parameters of the model, such that the model is able to efficiently master unseen tasks with a little amount of data from the tasks. The proposed algorithm makes it possible to separate task optimization and task identification, specifically, the model-based learner helps to identify the pattern of a task, while the gradient-based metalearner is capable of consistently improving the performance with only a few gradient update steps through making use of the task embedding produced by the model-based learner. In addition, the choice of network for the model-based learner in the proposed method is also discussed, and the performance of networks with different depths is explored. Finally, the simulation results on reinforcement learning problems demonstrate that the proposed approach outperforms compared metalearning algorithms and delivers a new state-of-the-art performance on a variety of challenging control tasks.

Investigating potential molecular mechanisms of serum exosomal miRNAs in colorectal cancer based on bioinformatics analysis.

Colorectal cancer (CRC) is the most common malignant gastrointestinal tumor worldwide. Serum exosomal microRNAs (miRNAs) play a critical role in tumor progression and metastasis. However, the underlying molecular mechanisms are poorly understood.The miRNAs expression profile (GSE39833) was downloaded from Gene Expression Omnibus (GEO) database. GEO2R was applied to screen the differentially expressed miRNAs (DEmiRNAs) between healthy and CRC serum exosome samples. The target genes of DEmiRNAs were predicted by starBase v3.0 online tool. The gene ontology (GO) and Kyoto Encyclopedia of Genomes pathway (KEGG) enrichment analysis were performed using the Database for Annotation, Visualization and Integrated Discovery (DAVID) online tool. The protein-protein interaction (PPI) network was established by the Search Tool for the Retrieval of Interacting Genes (STRING) visualized using Cytoscape software. Molecular Complex Detection (MCODE) and cytohubba plug-in were used to screen hub genes and gene modules.In total, 102 DEmiRNAs were identified including 67 upregulated and 35 downregulated DEmiRNAs, and 1437 target genes were predicted. GO analysis showed target genes of upregulated DEmiRNAs were significantly enriched in transcription regulation, protein binding, and ubiquitin protein ligase activity. While the target genes of downregulated DEmiRNAs were mainly involved in transcription from RNA polymerase II promoter, SMAD binding, and DNA binding. The KEGG pathway enrichment analyses showed target genes of upregulated DEmiRNAs were significantly enriched in proteoglycans in cancer, microRNAs in cancer, and phosphatidylinositol-3 kinases/Akt (PI3K-Akt) signaling pathway, while target genes of downregulated DEmiRNAs were mainly enriched in transforming growth factor-beta (TGF-beta) signaling pathway and proteoglycans in cancer. The genes of the top 3 modules were mainly enriched in ubiquitin mediated proteolysis, spliceosome, and mRNA surveillance pathway. According to the cytohubba plugin, 37 hub genes were selected, and 4 hub genes including phosphoinositide-3-kinase regulatory subunit 1 (PIK3R1), SRC, cell division cycle 42 (CDC42), E1A binding protein p300 (EP300) were identified by combining 8 ranked methods of cytohubba.The study provides a comprehensive analysis of exosomal DEmiRNAs and target genes regulatory network in CRC, which can better understand the roles of exosomal miRNAs in the development of CRC. However, these findings require further experimental validation in future studies.

Evaluation of genetic diversity and construction of DNA fingerprinting in Polygonatum Mill. based on EST-SSR and SRAP molecular markers.

Polygonatum sibiricum is widely consumed as a traditional Chinese herb and edible plant in China. Despite its nutritional and medical values, research on Polygonatum Mill. has been scarce, particularly as far as its genetic diversity is concerned. In this study, fourteen expressed sequence tag-derived simple sequence repeat (EST-SSR) and seven sequence-related amplified polymorphism (SRAP) markers were used to evaluate the genetic diversity in fifty Polygonatum Mill. accessions. The EST-SSRs and SRAPs produced 173 (90.58%) and 113 (93.39%) polymorphic bands, respectively. Unweighted Pair-Group Method Analysis (UPGMA) based on the combined data matrices of EST-SSRs and SRAPs divided the fifty Polygonatum Mill. accessions into fourteen groups. In addition, accessions of P. cyrtonema Hua obtained from Anhui and Zhejiang provinces were clustered according to their geographic origin. Furthermore, some accessions were gathered together based on species, such as P. kingianum Coll. et Hemsl, P. punctatum Royle ex Kunth, P. odoratum (Mill.) Druce, and P. sibiricum Red., and bootstrap analysis for clustering fully supported the grouping of the accessions. The Analysis of Molecular Variance (AMOVA) results revealed higher variation within populations (95%) rather than among populations (5%), indicating that Polygonatum Mill. has a low genetic differentiation between populations, and Principal Coordinate Analysis (PCoA) greatly supported the results of cluster analysis and AMOVA analysis. Finally, five markers which could produce abundant and stable bands were used to construct DNA fingerprinting database of Polygonatum Mill.. Our results demonstrated the utility of both EST-SSR and SRAP markers to successfully evaluate and identify Polygonatum Mill..

Short-term effects of meteorological factors on hand, foot and mouth disease among children in Shenzhen, China: Non-linearity, threshold and interaction.

BACKGROUND: Various meteorological factors have been associated with hand, foot and mouth disease (HFMD) among children; however, fewer studies have examined the non-linearity and interaction among the meteorological factors. METHODS: A generalized additive model with a log link allowing Poisson auto-regression and over-dispersion was applied to investigate the short-term effects daily meteorological factors on children HFMD with adjustment of potential confounding factors. RESULTS: We found positive effects of mean temperature and wind speed, the excess relative risk (ERR) was 2.75% (95% CI: 1.98%, 3.53%) for one degree increase in daily mean temperature on lag day 6, and 3.93% (95% CI: 2.16% to 5.73%) for 1m/s increase in wind speed on lag day 3. We found a non-linear effect of relative humidity with thresholds with the low threshold at 45% and high threshold at 85%, within which there was positive effect, the ERR was 1.06% (95% CI: 0.85% to 1.27%) for 1 percent increase in relative humidity on lag day 5. No significant effect was observed for rainfall and sunshine duration. For the interactive effects, we found a weak additive interaction between mean temperature and relative humidity, and slightly antagonistic interaction between mean temperature and wind speed, and between relative humidity and wind speed in the additive models, but the interactions were not statistically significant. CONCLUSIONS: This study suggests that mean temperature, relative humidity and wind speed might be risk factors of children HFMD in Shenzhen, and the interaction analysis indicates that these meteorological factors might have played their roles individually.

[THz time-domain spectroscopic study of composites filled with different carbon black].

Composites were prepared by filling high density polyethylene (HDPE) with acetylene black (AC-CB) and high-structure CB (HG-CB), respectively. Optical properties of the composites were characterized with terahertz time-domain spectroscopy (THz-TDS). It was found that as frequency increases the absorption coefficients of the composites increase whereas the refractive indexes decrease. Both the absorption coefficient and refractive index increase with increasing the particle concentration. The HG-CB filled composites have larger absorption coefficient but smaller refractive index compared with that of the AC-CB composites at the same particle concentration. These phenomena are related to the different particulate structures and aggregate structures of the CB particles. Assuming that the dielectric loss in THz frequency range is mainly attributed to the electron transport within the conductive clusters and the interfacial polarization of HDPE, the information of relaxation time and relaxation strength was obtained through fitting the experimental results to two-Debye theory of dipole relaxation.

Wet chemical synthesis and magnetic properties of single crystal Co nanochains with surface amorphous passivation Co layers.

: In this study, for the first time, high-yield chain-like one-dimensional (1D) Co nanostructures without any impurity have been produced by means of a solution dispersion approach under permanent-magnet. Size, morphology, component, and structure of the as-made samples have been confirmed by several techniques, and nanochains (NCs) with diameter of approximately 60 nm consisting of single-crystalline Co and amorphous Co-capped layer (about 3 nm) have been materialized. The as-synthesized Co samples do not include any other adulterants. The high-quality NC growth mechanism is proposed to be driven by magnetostatic interaction because NC can be reorganized under a weak magnetic field. Room-temperature-enhanced coercivity of NCs was observed, which is considered to have potential applications in spin filtering, high density magnetic recording, and nanosensors. PACS: 61.46.Df; 75.50; 81.07.Vb; 81.07.

A multiphasic model for the volume change of polyelectrolyte hydrogels.

A multiphasic model for the volume change of polyelectrolyte hydrogels that takes into account conservation of mass and momentum is derived. The gradient of chemical/electrochemical potentials of water and mobile ions is taken as the driving force for the volume change of the polyelectrolyte hydrogel, which is damped by the frictional forces between different phases and balanced by the elastic restoring force of the polymer network. Employing the model constructed here, the free swelling of a spherical polyelectrolyte hydrogel immersed in salt solution is simulated by the finite element method. The simulation shows that the polyelectrolyte hydrogel swells from the surface to the interior when the concentration of the external salt solution decreases. The swelling kinetics for ordinary hydrogels with high frictional coefficient between the polymer network and water is controlled by the collective diffusion of the polymer network, while for fast-response hydrogels it is controlled by the ionic diffusion in the hydrogel.