University of Electronic Science and Technology of China

About: University of Electronic Science and Technology of China is a education organization based out in Chengdu, China. It is known for research contribution in the topics: Antenna (radio) & Dielectric. The organization has 50594 authors who have published 58502 publications receiving 711188 citations. The organization is also known as: UESTC.

CA-Net: Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation

Abstract: Accurate medical image segmentation is essential for diagnosis and treatment planning of diseases. Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance for automatic medical image segmentation. However, they are still challenged by complicated conditions where the segmentation target has large variations of position, shape and scale, and existing CNNs have a poor explainability that limits their application to clinical decisions. In this work, we make extensive use of multiple attentions in a CNN architecture and propose a comprehensive attention-based CNN (CA-Net) for more accurate and explainable medical image segmentation that is aware of the most important spatial positions, channels and scales at the same time. In particular, we first propose a joint spatial attention module to make the network focus more on the foreground region. Then, a novel channel attention module is proposed to adaptively recalibrate channel-wise feature responses and highlight the most relevant feature channels. Also, we propose a scale attention module implicitly emphasizing the most salient feature maps among multiple scales so that the CNN is adaptive to the size of an object. Extensive experiments on skin lesion segmentation from ISIC 2018 and multi-class segmentation of fetal MRI found that our proposed CA-Net significantly improved the average segmentation Dice score from 87.77% to 92.08% for skin lesion, 84.79% to 87.08% for the placenta and 93.20% to 95.88% for the fetal brain respectively compared with U-Net. It reduced the model size to around 15 times smaller with close or even better accuracy compared with state-of-the-art DeepLabv3+. In addition, it has a much higher explainability than existing networks by visualizing the attention weight maps. Our code is available at https://github.com/HiLab-git/CA-Net .

High-Performance Ultraviolet Photodetector Based on a Few-Layered 2D NiPS3 Nanosheet

Abstract: 2D materials, represented by transition metal dichalcogenides (TMDs), have attracted tremendous research interests in photoelectronic and electronic devices. However, for their relatively small bandgap (<2 eV), the application of traditional TMDs into solar-blind ultraviolet (UV) photodetection is restricted. Here, for the first time, NiPS3 nanosheets are grown via chemical vapor deposition method. The nanosheets thinning to 3.2 nm with the lateral size of dozens of micrometers are acquired. Based on the various nanosheets, a linearity is found between the Raman intensity of specific Ag modes and the thickness, providing a convenient method to determine their layer numbers. Furthermore, a UV photodetector is fabricated using few-layered 2D NiPS3 nanosheets. It shows an ultrafast rise time shorter than 5 ms with an ultralow dark current less than 10 fA. Notably, this UV photodetector demonstrates a high detectivity of 1.22 × 1012 Jones, outperforming some traditional wide-bandgap UV detectors. The wavelength-dependent photoresponsivity measurement allows the direct observation of an admirable cut-off wavelength at 360 nm, which indicates a superior spectral selectivity. The promising photodetector performance, accompanied with the controllable fabrication and transfer process of nanosheet, lays the foundation of applying 2D semiconductors for ultrafast UV light detection.

Synergy of Big Data and 5G Wireless Networks: Opportunities, Approaches, and Challenges

Abstract: This article presents the synergistic and complementary features of big data and 5G wireless networks. An overview of their interplay is provided first, including big-data-driven networking and big data assisted networking. The former exploits heterogeneous resources such as communication, caching, and computing in 5G wireless networks to support big data applications and services, by catering for big data's features such as volume, velocity, and variety. The latter leverages big data techniques to collect wireless big data and extract in-depth knowledge regarding the networks and users to improve network planning and operation. To further illustrate the mutual benefits, two case studies on network aided data acquisition and big data assisted edge content caching are provided. Finally, some interesting open research issues are discussed.

Neural Network Control of a Rehabilitation Robot by State and Output Feedback

Abstract: In this paper, neural network control is presented for a rehabilitation robot with unknown system dynamics. To deal with the system uncertainties and improve the system robustness, adaptive neural networks are used to approximate the unknown model of the robot and adapt interactions between the robot and the patient. Both full state feedback control and output feedback control are considered in this paper. With the proposed control, uniform ultimate boundedness of the closed loop system is achieved in the context of Lyapunov's stability theory and its associated techniques. The state of the system is proven to converge to a small neighborhood of zero by appropriately choosing design parameters. Extensive simulations for a rehabilitation robot with constraints are carried out to illustrate the effectiveness of the proposed control.

Survey on the Internet of Vehicles: Network Architectures and Applications

Abstract: The vehicular ad hoc network (VANET) has been widely used as an application of mobile ad hoc networking in the automotive industry. However, in the 5G/B5G era, the Internet of Things as a cutting-edge technology is gradually transforming the current Internet into a fully integrated future Internet. At the same time, it will promote the existing research fields to develop in new directions, such as smart home, smart community, smart health, and intelligent transportation. The VANET needs to accelerate the pace of technological transformation when it has to meet the application requirements of intelligent transportation systems, vehicle automatic control, and intelligent road information service. Based on this context, the Internet of Vehicles (IoV) has come into being, which aims to realize the information exchange between the vehicle and all entities that may be related to it. IoV's goals are to reduce accidents, ease traffic congestion, and provide other information services. At present, IoV has attracted much attention from academia and industry. In order to provide assistance to relevant research, this article designs a new network architecture for the future network with greater data throughput, lower latency, higher security, and massive connectivity. Furthermore, this article explores a comprehensive literature review of the basic information of IoV, including basic VANET technology, several network architectures, and typical application of IoV.