Search

With the Internet of Vehicles (IOV), a lot of self-driving vehicles (SDVs) need to handle a variety of tasks but have very seriously limited computing and storage resources, meaning they cannot complete intensive tasks timely. In this paper, a joint edge computing and caching based on a Dueling Double Deep Q Network (D3QN) is proposed to solve the problem of the multi-task joint edge calculation and caching process. Firstly, the processes of offloading tasks and caching them to the base station are modeled as optimization problems to maximize system revenues, which are limited by system latency and energy consumption as well as cache space for computing task constraints. Moreover, we also take into account the negative impact of the number of unfinished tasks in relation to the optimization problem—the higher the number of unfinished tasks, the lower the system revenue. Secondly, we use the D3QN algorithm together with the cache models to solve the formulated NP-hard problem and select the optimal caching and offloading action by adopting an e-greedy strategy. Moreover, two cache models are proposed in this paper to cache tasks, namely the active cache, based on the popularity of the task, and passive cache, based on the D3QN algorithm. Additionally, tasks which deal with cache space are updated by computing the expulsion value based on type of popularity. Finally, simulation results show that the proposed algorithm has good performance in terms of the latency and energy consumption of the system and that it improves utilization of cache space and reduces the probability of unfinished tasks. Compared to the Deep Q Network with caching policy, with the Double Deep Q Network with caching policy and Dueling Deep Q Network with caching policy, the system revenue of the proposed algorithm is improved by 65%, 35% and 66%, respectively. The scenario of the IOV proposed in this article can be expanded to larger-scale IOV systems by increasing the number of SDVs and base stations, and the content caching and download functions of the Internet of Things can also be achieved through collaboration between multiple base stations. However, only the cache model is focused on in this article, and the design of the replacement model is not good enough, resulting in a low utilization of cache resources. In future work, we will analyze how to make joint decisions based on multi-agent collaboration for caching, offloading and replacement in IOV scenarios with multiple heterogeneous services to support different Vehicle-to-Everything services.

Mucosal healing, the result of endoscopic remission, is associated with prolonged clinical remission and delayed deterioration of Crohn's disease, which is significant and accompanied by reduced hospitalizations and surgeries. Currently, the relationship between ultrasonic parameters and mucosal healing remains controversial. To establish an ultrasonic regression model to evaluate mucosal healing, we conducted this preliminary study using multiple parameters from B-mode ultrasonography, colour Doppler flow imaging and shear wave elastography systematically.This study consisted of two single-centre investigations based on development and validation populations who received endoscopies (as the gold standard) and ultrasound. The involved bowel segments were divided into mucosal healing (MH) and nonmucosal healing (NMH) groups according to endoscopic results. Eight ultrasonic parameters were observed, including bowel wall thickness (BWT), mesenteric fat thickness (MFT), median modulus of elasticity (Emean), average shear wave velocity (SWV), Limberg scoring (LG), bowel wall stratification (BWS), ascites (AS) and lymph node enlargement (LN). We developed an ultrasonic regression model in the development phase to evaluate segmental mucosal healing and undertook prospective validation of this model.A total of 124 patients with 380 involved bowel segments from the development and validation cohorts were evaluated. Eight ultrasonic parameters were significantly different between the two groups (P0.05) in the development phase. Four significant parameters with better AUC performance were selected to establish an ultrasonic regression model to predict mucosal healing. The AUCs of this ultrasonic model were 0.975 and 0.942 in the development and validation cohorts, respectively.The multimodal ultrasonic model has the potential to evaluate segmental mucosal healing in Crohn's disease.

In this article, a predictive position control method based on a novel input-constrained nonlinear dynamic model (NDM) is proposed for time-varying position tracking of planar motors. The motivation lies in the possible utility of this method for motion systems. This method uses NDM subject to input constraint to deal with actuator saturation rather than uses a constrained optimization problem, such that it differs from conventional model predictive control. The NDM is represented in state–space equations (SSEs) to describe dynamic behaviors of the system constituted by the planar motor and an input saturation module. In contrast to linear SSEs, this model has the same linear vector-matrix form; the difference is that it applies saturation functions of states to replace states of state equation in linear SSEs for representing nonlinearity. By employing a self-designed neural network, the parameters of this model are determined via experimental sample data. With this model, a nonlinear multistep predictive model subject to input constraint is developed. Additionally, an explicitly analytical state feedback control law is approximately deduced by solving an unconstrained optimization problem subject to the nonlinear predictive model. Finally, simulation and experimental results show the effectiveness of the proposed method.

Background Elastography has not been widely applied to the gastrointestinal tract. The bowel wall's normal elasticity values are still unknown and are necessary for studies of gastrointestinal diseases. This study explores the feasibility of using shear wave elastography (SWE) to measure the terminal ileum wall stiffness in healthy subjects and establish the corresponding normal ranges of elasticity values. Methods This observational study recruited 139 healthy adult volunteers from April to July 2020. All examinations were performed in the anterior terminal ileum wall. Shear wave velocity (SWV) and Young's modulus (E) values were measured in the midline on longitudinal sections and replicated different operators' obtained data. Also, bowel wall thickness (BWT) and depth were recorded. Subgroups were classified according to the volunteers' gender, age, body mass index (BMI), BWT, and depth. The intra-class correlation coefficient was calculated to analyze inter- and intra-operator consistency, and independent t-tests and one-way analysis of variance were used to explore the differences in variables. Results The inter- and intra-operator agreements were good to excellent by different operators and in the replicated measurements (intra-operator consistency: 0.963; inter-operator consistency: 0.842). In all volunteers, the mean SWV was 1.08±0.25 m/s, the mean E value was 3.84±1.84 kPa, and the median BWT was 2 mm. SWV and E did not show significant differences according to gender (P=0.589), age (P=0.738), BMI (P=0.678), depth (P=0.375), or BWT (P=0.410). BWT did not show significant differences according to age (P=0.142), BMI (P=0.863), or depth (P=0.368). Conclusions SWE can be used in terminal ileum wall stiffness measurements with good reliability, and the SWE values do not appear to vary significantly according to different physiological factors. The corresponding elasticity ranges of the terminal ileum in normal adults were acquired.

In this study, a hybrid steady-state compressor model is proposed that can be used in the real-time performance monitoring, control and optimization of the vapor compression cycle. In the proposed model, first, a detailed analysis of the mass flow rate is presented, which is based on the volumetric efficiency concept and the assumption of a polytropic compression process. Then, discharge temperature of the refrigerant and power consumption of the compressor are also investigated. Three semiempirical models are constructed respectively. Further, to tune the unknown empirical parameters of the models, a social learning particle swarm optimization (SLPSO) algorithm is developed by using the real-time experimental data. An experimental apparatus of a refrigerant system is tested to validate the proposed models. The experimental results demonstrate that the proposed models accurately predict the performance of real-time operating compressors. Meanwhile, the models identified by the SLPSO algorithm are more accurate than those identified by the traditional least-squares method.

The existing real-time object detection algorithm often omits the objects in the object detection. So an improved Tiny YOLOv3 (you look only once) algorithm is proposed with both lightweight and high accuracy of object detection. The improved Tiny YOLOv3 uses K-means clustering to estimate the size of the anchor boxes for dataset. The pooling and convolution layers are added in the network to strengthen feature fusion and reduce parameters. The network structure increases upsampling and downsampling to enhance multi-scale fusion. The complete intersection over union is added in the loss function, which effectively improves the detection results. In addition, the proposed method has the lightweight module size and can be trained in the CPU. The experimental results show that the proposed method can meet the requirements of the detection speed and accuracy.

BackgroundPrevious studies have suggested that pregnant women with pulmonary hypertension (PH) have high maternal mortality. However, indexes or factors that can predict maternal death are lacking.MethodsWe retrospectively reviewed pregnant women with PH admitted for delivery from 2012 to 2020 and followed them for over 6 months. The patients were divided into two groups according to 10-day survival status after delivery. Predictive models and predictors for maternal death were identified using four machine learning algorithms: naïve Bayes, random forest, gradient boosting decision tree (GBDT), and support vector machine.ResultsA total of 299 patients were included. The most frequent PH classifications were Group 1 PH (73.9%) and Group 2 PH (23.7%). The mortality within 10 days after delivery was 9.4% and higher in Group 1 PH than in the other PH groups (11.7 vs. 2.6%, P = 0.016). We identified 17 predictors, each with a P-value < 0.05 by univariable analysis, that were associated with an increased risk of death, and the most notable were pulmonary artery systolic pressure (PASP), platelet count, red cell distribution width, N-terminal brain natriuretic peptide (NT-proBNP), and albumin (all P < 0.01). Four prediction models were established using the candidate variables, and the GBDT model showed the best performance (F1-score = 66.7%, area under the curve = 0.93). Feature importance showed that the three most important predictors were NT-proBNP, PASP, and albumin.ConclusionMortality remained high, particularly in Group 1 PH. Our study shows that NT-proBNP, PASP, and albumin are the most important predictors of maternal death in the GBDT model. These findings may help clinicians provide better advice regarding fertility for women with PH.

To tackle the problem of the desired signal (DS) steering vector mismatch, especially in the situation of direction-of-arrival error and array perturbations, a robust interference-plus-noise covariance matrix (INCM) reconstruction method based upon DS removal is presented. Unlike previous studies, this paper proposes to remove the DS component from the training data by building a blocking matrix, which is computed as the inverse of the DS-plus-noise covariance matrix (DSNCM). More specifically, to increase the robustness against arbitrary mismatches, the DS steering vector estimated as the prime eigenvector of the DS matrix, which is attained through integrating the Capon spectrum estimator over the annulus uncertainty sets of the mainlobe region in advance, is adopted to give a faithful blocking matrix. After that, utilizing the obtained blocking matrix to process the training data, the quasi INCM is computed indeed. Finally, a precise INCM is reconstructed by projecting the principal components of the quasi INCM onto the aforesaid DSNCM. Numerical simulations have illustrated that the proposed adaptive beamformer can outperform the existing ones and gain almost optimal performance under different scenarios.

Background: To study the effects of long non-coding ribonucleic acid (lncRNA) HOX transcript antisense intergenic RNA (HOTAIR) on the proliferation and apoptosis of malignant melanoma cells, and to explore its specific regulatory mechanism through the nuclear factor-κB (NF-κB) signaling pathway. Methods: LncRNA HOTAIR small-interfering RNAs (siRNAs) were designed and synthesized, and the effects of si-HOTAIR transfection on the proliferation and apoptosis of malignant melanoma cells were detected via cell counting kit-8 (CCK-8) assay, 4',6-diamidino-2-phenylindole (DAPI) staining assay and flow cytometry, respectively. The gene expressions were determined using quantitative reverse transcription-polymerase chain reaction (qRT-PCR), the changes in NF-κB pathway-related proteins and apoptosis-associated proteins after interference in lncRNA HOTAIR were detected via Western blotting, and the level of NF-κB in each group was determined via ELISA. Results: The results of CCK-8 assay revealed that the cell proliferation rate significantly declined gradually in si-HOTAIR group compared with that in si-NC group and control group (P

Purpose. OSA is closely associated with hypertension, and both epidemiological and experimental studies have confirmed that OSA is one of the most important independent risk factors for hypertension. The pathological mechanisms by which OSA causes hypertension are not well understood, and in this paper, we explored the molecular mechanisms by which OSA may mediate hypertension at the bioinformatics level. Materials and Methods. We downloaded disease-related datasets from the GEO public database, calculated the differential genes between the two groups of patients by the limma package, and then further constructed the WGCNA network based on the clinical characteristics of patients to explore the important regulatory genes in the disease. Subsequently, ssGSEA was used to explore the potential molecular mechanisms of disease progression and GSVA was applied to analyze the specific signaling pathways. Finally, we performed real-time quantitative PCR (qRT-PCR) to validate these pivotal genes. Results. Three genes were selected as target genes, namely, GPR179, RNF150, and JPH4. The results showed they were strongly correlated with immune cell content that high expression of the three core genes was associated with myogenesis, angiogenesis, oxidation, metabolism, and PI3K/AKT/mTOR pathways. qRT-PCR validated that all three genes have statistically significant differences between the OSA group and OSA combined with hypertension group. Conclusion. Our study provides new evidence for the potential molecular mechanisms of OSA combined with hypertensive disease as well as diagnosis and treatment.

This paper develops a Sobol sensitivity analysis model of thrust force of linear switched reluctance motors (LSRMs), to quantitatively analyze the coupled influence of parameters on the thrust force ripple. A nonlinear thrust force model of a developed LSRM is first given. Using Sobol sensitivity analysis method, the local and global sensitivity indexes are deduced according to thrust force model. In addition, applying the deduced sensitivity indexes, local and global sensitivity results of thrust force are obtained, and the quantitatively coupled influence of parameters on the thrust force ripple is presented and analyzed.

The identification of dynamic model parameters of the permanent magnet synchronous linear motors (PMSLMs) is the foundation of high-performance motor control. To address the low accuracy of modelling of PMSLMs, a parameter identification method of the PMSLM is proposed by using multi-innovation least squares (MILS) approach in this paper. The linear transfer function of the PMSLM is first built. Then the MILS approach based on the recursive least squares (RLS) algorithm is developed to identify the parameters of the built model. Finally, the simulation results of MILS and RLS algorithms are shown, and the effectiveness of the proposed method is verified.

Increasing the allowable gas pressure of underground gas storage (UGS) is one of the most effective methods to increase its working gas capacity. In this context, hydraulic fracturing tests are implemented on the target formation for the UGS construction of Jintan salt caverns, China, in order to obtain the minimum principal in situ stress and the fracture breakdown pressure. Based on the test results, the maximum allowable gas pressure of the Jintan UGS salt cavern is calibrated. To determine the maximum allowable gas pressure, KING-1 and KING-2 caverns are used as examples. A three-dimensional (3D) geomechanical model is established based on the sonar data of the two caverns with respect to the features of the target formation. New criteria for evaluating gas penetration failure and gas seepage are proposed. Results show that the maximum allowable gas pressure of the Jintan UGS salt cavern can be increased from 17 MPa to 18 MPa (i.e. a gradient of about 18 kPa/m at the casing shoe depth). Based on numerical results, a field test with increasing maximum gas pressure to 18 MPa has been carried out in KING-1 cavern. Microseismic monitoring has been conducted during the test to evaluate the safety of the rock mass around the cavern. Field monitoring data show that KING-1 cavern is safe globally when the maximum gas pressure is increased from 17 MPa to 18 MPa. This shows that the geomechanical model and criteria proposed in this context for evaluating the maximum allowable gas pressure are reliable. Keywords: Underground gas storage (UGS) salt cavern, In situ stress testing, Maximum gas pressure, Gas penetration failure, Microseismic monitoring

This paper presents an online air balancing method applied in dedicated outdoor air systems, named as gradient-based online adaptive balancing (GOAB) method. The GOAB method derives the damper adjustment rule based on the gradient of the objective function that quantifies the discrepancy between actual flow rate and the set point in normalized form. An adaptive mechanism is applied to capture the change of Jacobian matrix during the balancing process. The optimum direction of damper angle increment is determined based on the null space of the gradient vector to speed up the convergence and ensure at least one damper to be fully open. Compared with the existing methods, the proposed GOAB method offers the following advantages:1.) The method can be performed on a duct system in operation, which provides possibility to re-balance the system under dynamic working conditions; 2.) The energy loss on dampers is minimized without compromising the balancing accuracy. Simulations are performed in MATLAB to investigate several model parameters and provide guidance for the parameter design. Experiments are conducted on a 5-ternimal duct system to validate the proposed method under various design flow requirements. The results show that the maximum absolute percentage error can be controlled within 4.3% among all test cases. This method improves the energy efficiency of the duct system and speeds up the convergence rate towards air balance to save time and cost. Ministry of Education (MOE) This work is supported by Tier 1 project: Optimization and Control for Fan Array Based Air Handling Units (Grant Number: 2017-T1-001- 083 (RG 88/17)) by Ministry of Education Singapore.