Search

BackgroundThere is no standard neoadjuvant therapy for locally advanced esophageal cancer in China. The role of neoadjuvant chemotherapy plus immunotherapy for locally advanced esophageal cancer is still being explored.MethodsThis open-label, randomized phase II study was conducted at a single center between July 2019 and September 2020; 30 patients with locally advanced esophageal squamous cell carcinoma (ESCC) (T3, T4, or lymph-node positive) were enrolled. Patients were randomized according to the enrollment order at a 1:1 ratio to receive chemotherapy on day 1 and toripalimab on day 3 (experimental group) or chemotherapy and toripalimab on day 1 (control group). The chemotherapeutic regimen was paclitaxel and cisplatin. Surgery was performed 4 to 6 weeks after the second cycle of chemoimmunotherapy. The primary endpoint was pathological complete response (pCR) rate, and the secondary endpoint was safety and disease-free survival.ResultsThirty patients completed at least one cycle of chemoimmunotherapy; 11 in the experimental group and 13 in the control group received surgery. R0 resection was performed in all these 24 patients. Four patients (36%) in the experimental group and one (7%) in the control group achieved pCR. The experimental group showed a statistically non-significant higher pCR rate (p = 0.079). PD-L1 combined positive score (CPS) examination was performed in 14 patients; one in the control group had a PD-L1 CPS of 10, and pCR was achieved; the remaining 13 all had ≤1, and 11 of the 13 patients received surgery in which two (in the experimental group) achieved pCR. Two patients endured ≥grade 3 adverse events, and one suffered from grade 3 immune-related enteritis after one cycle of chemoimmunotherapy and dropped off the study. Another patient died from severe pulmonary infection and troponin elevation after surgery.ConclusionsAlthough the primary endpoint was not met, the initial results of this study showed that delaying toripalimab to day 3 in chemoimmunotherapy might achieve a higher pCR rate than that on the same day, and further large-sample clinical trials are needed to verify this.Clinical Trial RegistrationClinicalTrials.gov, identifier NCT 03985670.

Background There is no standard neoadjuvant therapy for locally advanced esophageal cancer in China. The role of neoadjuvant chemotherapy plus immunotherapy for locally advanced esophageal cancer is still being explored. Methods This open-label, randomized phase II study was conducted at a single center between July 2019 and September 2020, 30 patients with locally advanced ESCC (T3, T4, or lymph-node positive) were enrolled. Patients were randomized according to the enrollment order at a 1:1 ratio to receive chemotherapy on day 1 and toripalimab on day 3 (experiment group) or chemotherapy and toripalimab on day 1 (control group). The chemotherapeutic regimen was paclitaxel and cisplatin. Surgery was performed 4 to 6 weeks after the second cycle of chemoimmunotherapy. The primary endpoint was pCR rate, the secondary endpoint were safety and disease-free survival. Results Thirty patients completed at least one cycle of chemoimmunotherapy; 11 in the experimental group and 13 in the control group received surgery. R0 resection was got in all these 24 patients. Four patients (36%) in the experimental group and one (7%) in the control group achieved pCR. The experimental group showed a statistically non-significant higher pCR rate (P = 0.079). PD-L1 combined positive score (CPS) examination was performed in 14 patients; one in control group had a PD-L1 CPS of 10 and pCR was achieved, while the left 13 were all ≤ 1, 11 of the 13 patients received surgery in which two (in the experimental group) got pCR. Two patients endured from ≥ grade 3 adverse events, one suffered from grade 3 immune-related enteritis after one cycle of chemoimmunotherapy and dropped off the study. Another patient died from severe pulmonary infection and troponin elevation after surgery. CONCLUSIONS Although the primary endpoint was not met, initial results of this study showed that delaying toripalimab to day 3 in chemoimmunotherapy might achieve a higher pCR rate than that on same day, and further large-sample clinical trials are needed to verify this. Trial registration: NCT 03985670, registered 10 June 2019.

Media-Based Modulation (MBM) is regarded as a promising technique for future massive machine-type communications (mMTC) due to its high energy/spectral efficiency, good error performance and low-complexity radio frequency hardware implementation. In this paper, we consider both sparsity nature of user activity and sparsity nature of MBM signals in the uplink MBM-enabled mMTC system. According to the static user activation or the dynamic user activation in a coherent time, we classify the transmission schemes into two types and propose corresponding improved compressive sensing (CS)-based joint user identification and data detection with/without prior information of channel state information (CSI). The simulation results demonstrate the performance advantages of our proposed algorithms over the state-of-the-art CS-based user detection methods or CS-based symbol detection methods and evaluate the performance with different system parameters.

Background Transcriptome data generates massive amounts of information that can be used for characterization and prognosis of patient outcomes for many diseases. The goal of our research is to predict the survival time of lung adenocarcinoma patients and improve the accuracy of classifying the long-survival cohort and short-survival cohort. Methods We filtered prognostic features related with survival time of lung adenocarcinoma patients by the method of Relief and predicted whether survival time of the patient is >3 years or not—using eight machine learning algorithms (Support Vector Machines, Random Forests, Logistic Regression, Naïve Bayes, Linear Regression, Support Vector Regression (kernel Poly), Support Vector Regression (kernel Linear), and Ridge Regression). Then the best-performed algorithm was chosen to build a predictive model of survival time of lung adenocarcinoma patients. Further, another dataset was used to verify the stability and suitability of this model. We explored the underlying mechanisms of RNA expression changes with the corresponding DNA mutations and DNA methylation patterns in the 22 selected genetic features. Results The best machine learning algorithm was Naïve Bayes (accuracy=75%, AUC =0.81) using the top 22 genetic features, and this algorithm had the stable and great performance on another dataset as well. The coupled mutation number of the long-survival group (>6 years) was less than the short-survival group (

This letter proposes a novel layered video multicast scheme for wireless-powered relay systems, where the multi-antenna base station multicasts layered video signals over non-orthogonal multiple access (NOMA) to heterogeneous user groups with the help of a dedicated relay. To minimize the average transmission power of the base station, an optimization problem is formulated to jointly design the energy signal power and layered video multicast beamformers, while guaranteeing all the users’ quality of experience (QoE), i.e., fluent video playback and successful decoding of video signals. Since the formulated optimization problem is non-convex, a low-complexity algorithm is developed to obtain a local minimum, and a novel initial feasible point search algorithm is proposed to initialize the solving process. Simulation results reveal that the proposed scheme works well and can greatly reduce the power consumption.

Spatial modulation (SM) with high spectrum and energy efficiencies is promising for future massive connectivity communication networks, in which user activities are often sporadic and SM signals are sparse. Such characteristics, if exploited in the detection process, will lead to gains in terms of receiver complexity and performance. This letter proposes a compressive sensing (CS) algorithm for multiuser detection in SM-MIMO systems to efficiently exploit such characteristics. Specifically, the temporal correlation of the user activities and the random transmission nature of SM signals are used to improve detection performance without requiring prior knowledge of the user sparsity level. In addition, the proposed algorithm requires a lower number of BS receive and transmit antennas in each user compared with existing CS-based detection algorithms and has a significantly lower computational complexity.

Machine (deep) learning-enabled accurate traffic modeling and prediction is an indispensable part for future big data-driven intelligent cellular networks, since it can help autonomic network control and management as well as service provisioning. Along this line, this paper proposes a novel deep learning architecture, namely Spatial–Temporal Cross-domain neural Network (STCNet), to effectively capture the complex patterns hidden in cellular data. By adopting a convolutional long short-term memory network as its subcomponent, STCNet has a strong ability in modeling spatial–temporal dependencies. Besides, three kinds of cross-domain datasets are actively collected and modeled by STCNet to capture the external factors that affect traffic generation. As diversity and similarity coexist among cellular traffic from different city functional zones, a clustering algorithm is put forward to segment city areas into different groups, and consequently, a successive inter-cluster transfer learning strategy is designed to enhance knowledge reuse. In addition, the knowledge transferring among different kinds of cellular traffic is also explored with the proposed STCNet model. The effectiveness of STCNet is validated through real-world cellular traffic datasets using three kinds of evaluation metrics. The experimental results demonstrate that STCNet outperforms the state-of-the-art algorithms. In particular, the transfer learning based on STCNet brings about 4%~13% extra performance improvements.

This paper investigates the energy efficient downlink transmission in distributed massive multiple-input–multiple-output systems, where massive antennas are clustered on geographically distributed remote radio heads (RRH). A closed form of the ergodic achievable rate is derived for a specific RRH-user association and imperfect channel state information to assist the formulation of its system energy efficiency (EE). By assuming minimum user achievable rate requirements and a practical power consumption model including both the transmit power and the antenna circuit power, an optimization problem is formulated to maximize the system EE with respect to the power allocation, the antennas activation, and the RRH-user association in a combinatorial manner. The formulated problem is a mixed-integer nonlinear programming problem, which is NP-hard and usually has no feasible solution. To solve it, an efficient algorithm is proposed by utilizing hierarchical decomposition technique and iterative successive convex approximation. The feasibility as well as the effectiveness of the proposed algorithm are analyzed and proved through analysis and intensive computer simulations. Results show that the system EE can be greatly improved by the proposed scheme.

In the modern machine tool manufacturing scene, as the milling tool of CNC machine tool, the health of the tool directly affects the processing efficiency and product quality. Effective prognostic and health management of tool is critical. Precise monitoring of tool wear helps to avoid product quality problems caused by tool fault and improve production efficiency. Therefore, this paper constructs a tool fault diagnosis method based on deep learning. In order to effectively fuse the vibration signals features from different directions of machine tool spindle, we apply a variety of Channel Attention (CA) mechanisms to Multiscale Network (MSNet) to construct Multiscale-Channel Attention Network (MS-CA Net), and explore the performance gains of these modules in tool wear classification tasks. Among these modules, the CA blocks include Channel Attention Block (CAB), Squeeze and Excitation Block (SEB), and Efficient Channel Attention Block (ECAB). We use the three proposed networks to classify the tool wear status by identifying the vibration signal of the machine tool spindle. At the same time, in order to verify the performance of the proposed tool fault diagnosis method in the actual milling scene, this paper designs a tool wear test platform to collect sample data that meets actual industrial production scenarios, which uses a three-axis accelerometer to collect tool life cycle vibration signals, and a digital universal tool microscope to measure tool wear values. The experimental results show that, compared with the MSNet method, the tool fault diagnosis accuracy rate of the improved method is increased by 4.47%-7.38%.

Personnel scheduling in assembly workshop is a kind of problem widely existing in machine tool manufacturing industry. In this paper, according to the actual production needs, to minimize the total assembly job completion time as the optimization objective, the mathematical model of assembly shop scheduling is established, and the Particle Swarm Optimization algorithm (PSO) is discretized. At the same time, we discretize the particle swarm optimization algorithm and introduce the mutation mechanism of Genetic Algorithm (GA) to propose a Hybrid Discrete Particle Swarm Optimization (HD-PSO) algorithm. In the actual production data of machine tool manufacturing enterprises, the algorithm proposed in this paper is compared with the standard PSO algorithm and GA. The results show that the algorithm proposed in this paper is more suitable for the needs of factory scheduling than other algorithms, and has better performance. In order to facilitate the use experience of enterprise employees, a Graphical User Interface (GUI) is designed to realize the intelligent personnel scheduling of assembly workshop.

With the development of the Industrial Internet of Things (IIoT) and the upgrading of the manufacturing industry, intelligent solutions for factories are imminent. Nowadays, most factories have backward testing methods and low efficiency, which has gradually become the shackles of enterprise development. Aiming at the problem of the appearance quality of the air conditioner external unit, this paper proposes an intelligent detection solution for the appearance quality of the air conditioner external unit based on the YOLO object detector, and constructs an image dataset containing with 25897 images and object labels. In the actual test, the object labelled in the quality inspection task of the air conditioner external unit reached a recall rate of more than 99% and a mAP of more than 95%.

Job shop scheduling problem (JSSP) is a NP-hard problem, which is widely used in manufacturing industry. In this paper, we proposed an improved hybrid Lion Swarm Optimization (IHLSO) algorithm for the JSSP based on the actual production requirements. The proposed IHLSO included two improved algorithms, which were respectively hybrid Lion Swarm Optimization (HLSO) algorithm and HLSO with Solis&Wets (SW) local search algorithm (SW-HLSO). IHLSO improved the global searching ability of basic Lion Swarm Optimization (LSO) algorithm by combing Particle Swarm Optimization (PSO) algorithm and improved the local searching ability by applying SW algorithm. Tests on the benchmark instances of the JSSP showed excellent performance of the improved LSO. In the actual production scene of machine tool manufacturing enterprises, the algorithm proposed in this paper is more suitable for the demand of factory scheduling than other methods. In addition, in order to facilitate the application of the algorithm in actual production and help the employees to arrange production more conveniently, we had designed a Graph User Interface (GUI) to realize intelligent production scheduling in digital workshop.

In this paper, we study a tethered unmanned aerial vehicle (TUAV) assisted cellular communication system, where the TUAV is deployed to provide downlink services for multiple ground users. The TUAV is linked and powered by an associated tether connected to the ground base station (GBS), which increases the hovering time and backhaul capacity. But the maximum tether length additionally limits the flying trajectory of TUAV. Taking this physical constraint into consideration, we jointly optimize the power allocation strategy and the trajectory design of the TUA V. To achieve fair performance among ground users, we maximize the minimum average throughput of all the users (also known as max-min throughput) over a given time period. Through analysis, we can conclude that the resulted optimization problem is non-convex and thus difficult to be solved optimally. To tackle this challenge, we resort to a successive convex optimization (SCO) technique to get an efficient suboptimal solution. Simulation results confirm that our proposed algorithm can achieve a substantially higher max-min throughput while guaranteeing fairness in comparison with other schemes.

With the continuous development of science and technology, the intelligent development of community system becomes a trend. Meanwhile, smart mobile devices and cloud computing technology are increasingly used in intelligent information systems; however, smart mobile devices such as smartphone and smart pad, also known as thin clients, limited by either their capacities (CPU, memory or battery) or their network resources, do not always meet users' satisfaction in using mobile services. Mobile cloud computing, in which resource-rich virtual machines of smart mobile device are provided to a customer as a service, can be terrific solution for expanding the limitation of real smart mobile device, but the resources utilization rate is low and the information cannot be shared easily. To address the problems above, this paper proposes an information system for intelligent community, which is composed of thin clients, wide band network and cloud computing servers. On one hand, the thin clients with the characteristics of energy efficiency, high robustness and high computing capacity can efficiently avoid the problems encountered in the PC architecture and mobile devices. On the other hand, the cloud computing servers in the proposed information system solve the problems of resource sharing barriers. Finally, the system is built in real environments to evaluate the performance. We deploy the proposed system in a community with more than 2000 residents, and it is demonstrated that the proposed system is robust and efficient.

Prognostic and health management is a key issue in the field of machine tool manufacturing. As the “teeth” of CNC machine tools, their health status directly affects the machining efficiency and the quality of products. Accurate monitoring of tool wear can help to avoid product quality problems caused by tool failure and improve productivity. In this paper, we investigate the deep learning-based tool fault diagnosis approach. First, a new data-driven tool fault diagnosis method based on improved multiscale network and feature fusion (IMSNet-F) is proposed to recognize and classify the tool wear condition. It can increase the efficiency of the process and make it possible to replace the tool before catastrophic wear occurs. And then, a tool wear experimental system is designed to verify the performance of the proposed tool fault diagnosis method in a real production scenario. Besides, based on the tool wear experimental system, a data set of vibration signals used to detect tool wear conditions is constructed and publicly released. Experimental results show the proposed approach can improve the tool fault diagnosis accuracy by 2.2% compared to existing methods.

With the continuous development of science and technology, the intelligent development of community system becomes a trend. Meanwhile, smart mobile devices and cloud computing technology are increasingly used in intelligent information systems; however, smart mobile devices such as smartphone and smart pad, also known as thin clients, limited by either their capacities (CPU, memory or battery) or their network resources, do not always meet users' satisfaction in using mobile services. Mobile cloud computing, in which resource-rich virtual machines of smart mobile device are provided to a customer as a service, can be terrific solution for expanding the limitation of real smart mobile device, but the resources utilization rate is low and the information cannot be shared easily. To address the problems above, this paper proposes an information system for intelligent community, which is composed of thin clients, wide band network and cloud computing servers. On one hand, the thin clients with the characteristics of energy efficiency, high robustness and high computing capacity can efficiently avoid the problems encountered in the PC architecture and mobile devices. On the other hand, the cloud computing servers in the proposed information system solve the problems of resource sharing barriers. Finally, the system is built in real environments to evaluate the performance. We deploy the proposed system in a community with more than 2000 residents, and it is demonstrated that the proposed system is robust and efficient.

With the development of Industrial Internet of Things (IIoT), the computation intensive tasks with restrict delay constraints generated at network edge emerge as the main challenge to the terminals with limited power and processing capability. The combination of edge and cloud computing has been demonstrated as one promising solution to such problem. In the edge-cloud collaboration (ECC) framework, the task scheduling among edges and cloud is of great importance on impacting the performance of the system. In this paper, we investigate the task offloading strategy to minimize the energy consumption of the networks. To achieve that, a time delay penalty mechanism, which searches the optimal power for edge to cloud task offloading under given delay constraint, is first proposed. On that basis, a low complexity edge-cloud matching algorithm leveraging the bipartite matching method is developed, to further minimize the execution energy consumption of all devices. Finally, to evaluate its efficiency, the proposed algorithm is deployed and tested on an novel edge-cloud computing collaboration platform. Both simulation and experiment results revel that our proposed scheme can achieve the less energy consumption compared with other alternatives. In addition, it also indicates that our proposed scheme can effectively matching resources from the edge to the cloud, especially for the issues that edge devices fail to meet demands due to limit processing ability.

Edge caching has been explored as an effective technology to alleviate the heavy traffic burden of the backhaul and avoid transmission congestion in vehicular networks. However, high mobility of vehicles could lead to repetitive content caching, resulting in high system cost. Because content popularity changes very frequently in vehicular networks, to provide better service for vehicle users, it is essential to update content frequently. This leads to expensive update cost at the same time. To reduce such cost, we propose a mobility-aware cost effective edge caching strategy, in which vehicle mobility, file encoding technology and dynamic content popularity are jointly taken into consideration. To reduce the complexity of formulated problem, deep reinforcement learning (DRL) approach is adopted. Simulation results show that the proposed mobility-aware coded edge caching strategy can dramatically reduce the system cost (up to 36% compared with classic caching algorithm).

For the purpose of solving the problem of water prevention for a vertical shaft passing through the sinian ultra-thick dolomite (SUTD) aquifer in Southwest China, the groutability of the SUTD and the technical parameters of the surface pre-grouting (SPG) were studied by a field grouting test and numerical simulation. Firstly, the field grouting test was conducted to investigate the groutability of the SUTD. According to the results of the field test, the SUTD could be divided into two members, and the upper member has greater groutability than that of the lower member. In the upper member, special measures should be utilized to avoid the overspread phenomenon, while a high pressure should be applied to increase the grout takes for ensuring the sealing effects in the lower member. Then, the rational borehole spacing and final grouting pressure were obtained through numerical simulation and the field SPG test. Finally, the effectiveness of the SPG was verified by the water pressure test. The residual water inflow was small enough to meet the shaft sinking after the SPG was completed, and the sealing efficiency was 98.04%. The results of the present paper could provide technical references for water controlling for shaft sinking under the similar geological conditions.

Aiming at the shortcomings of fuzzy C-means (FCM) clustering algorithm that it is easy to fall into local minima and sensitive to initial values and noisy data, this paper proposes a fuzzy clustering algorithm based on improved lion swarm optimization algorithm. Aiming at the problem that lion swarm optimization (LSO) algorithm is easy to fall into the local optimum, this paper improves lion swarm optimization algorithm by introducing sin cos algorithm and elite opposition-based learning. In addition, the introduction of a supervision mechanism enhances the lions’ ability to jump out of local optimum and improves the local search ability of lion swarm optimization algorithm. The optimal solution obtained by improved lion swarm optimization algorithm is used as the initial clustering center of FCM algorithm, then FCM algorithm is run to obtain the global optimal solution, which effectively overcomes the shortcomings of FCM algorithm. The experimental results show that, compared with original FCM clustering algorithm, FCM clustering algorithm based on improved lion swarm optimization algorithm has improved the algorithm’s optimization ability and has better clustering results.

Reconfigurable intelligent surface (RIS) provides a promising way to build the programmable wireless transmission environments in the future. Owing to the large number of reflecting elements used at the RIS, joint optimization for the active beamforming at the transmitter and the passive reflector at the RIS is usually complicated and time-consuming. To address this problem, this article proposes a low-complexity joint beamforming and reflecting algorithm based on fractional programing (FP). Specifically, we first consider a RIS-aided multi-user communication system with perfect channel state information (CSI) and formulate an optimization problem to maximize the sum rate of all users. Since the problem is nonconvex, we decompose the original problem into three disjoint subproblems. By introducing favorable auxiliary variables, we derive the closed-form expressions of the beamforming vectors and reflecting matrix in each subproblem, leading to a joint beamforming and reflecting algorithm with low complexity. We then extend our approach to handle the case when transmitter-RIS and RIS-receiver channels are not perfect and develop corresponding low-complexity joint beamforming and reflecting algorithm with practical channel estimation. Simulation results have verified the effectiveness of the proposed algorithms as compared to various benchmark schemes.

Real-time tool wear prediction is one of the key problems to be solved in the field of intelligent manufacturing. In recent years, with the help of powerful feature extraction capability of AutoEncoder(AE), the Stacked AutoEncoder-based Deep Neural Network(SAE-DNN) model has achieved good results on tool wear prediction. However, the SAE-DNN model ignores the time series characteristics between original samples. Besides, training the neural network with a gradient-based algorithm is prone to fall into local optimum. To solve the problems in the SAE-DNN, we propose a new improved model: Sequential Stacked AutoEncoder-based Artificial Neural Network(SSAE-ANN). First, the SSAE-ANN model uses the Sequential Stacked AutoEncoder(SSAE), which can not only fuse the features between different channels of a single sample, but also extract the time series features between adjacent samples. Second, the SSAE-ANN model uses the Improved Sheep Optimization(ISO) algorithm to train the neural network. Compared with the gradient algorithm, the ISO algorithm has stronger robustness and global optimization ability. At the end of the paper, we perform experiments on PHM2010 dataset and verify the superiority of SSAE-ANN model by comparing with other algorithms.