A5101900574- Reinforcement Learning in Robotics
- Medical Image Segmentation Techniques
- Web Data Mining and Analysis
- Artificial Intelligence in Games
- Antenna Design and Analysis
- Radiomics and Machine Learning in Medical Imaging
- Advanced Algorithms and Applications
- Image Retrieval and Classification Techniques
- Medical Imaging Techniques and Applications
- Human Motion and Animation
- Data Management and Algorithms
- Simulation and Modeling Applications
- Advanced Computational Techniques and Applications
- Adaptive Dynamic Programming Control
- Full-Duplex Wireless Communications
- Advanced Combustion Engine Technologies
- Target Tracking and Data Fusion in Sensor Networks
- Recommender Systems and Techniques
- Image and Object Detection Techniques
- Solidification and crystal growth phenomena
- Power Line Inspection Robots
- HIV Research and Treatment
- Machine Learning and ELM
- Evolutionary Algorithms and Applications
- Smart Grid Energy Management
Fuzhou University
2025
Zhejiang University
2006-2025
Quanzhou Institute of Equipment Manufacturing Haixi Institute
2025
Chinese Academy of Sciences
2024-2025
Fujian Institute of Research on the Structure of Matter
2024-2025
Shanghai Electric (China)
2022-2025
University of Chinese Academy of Sciences
2025
National University of Defense Technology
2022-2024
Shanghai Jiao Tong University
2022-2023
Ganzhou People's Hospital
2023
Antiretroviral therapy (ART) halts HIV-1 replication, reducing plasma virus levels to below the limit of detection, but it is not curative due a reservoir latently infected CD4+ T cells. In some people living with (PLWH), RNA becomes persistently detectable despite optimal ART. This nonsuppressible viremia (NSV) characterized by identical, non-evolving variants expressed from cell clones. The mechanisms driving persistent production specific population cells are poorly understood. We...
In this letter, a novel co-linearly polarized full-duplex antenna-in-package (AiP) with high Tx/Rx isolation and wide decoupled bandwidth is proposed for integrated sensing communication (ISAC). Fan-out wafer-level packaging technology employed, the preparation process discussed in details. To improve isolation, current direction of patch elements orthogonal to that feeding ports. Two identical 1 × 2 antenna arrays are used as Tx Rx antennas, respectively. further suppress mutual coupling,...
The popular deep<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M1"><mml:mrow><mml:mi>Q</mml:mi></mml:mrow></mml:math>learning algorithm is known to be instability because of the<mml:math id="M2"><mml:mrow><mml:mi>Q</mml:mi></mml:mrow></mml:math>-value’s shake and overestimation action values under certain conditions. These issues tend adversely affect their performance. In this paper, we develop the ensemble network architecture for deep reinforcement learning which based on...
MCAM, alternatively referred to as CD146, is an integral membrane glycoprotein belonging the immunoglobulin superfamily. However, its importance in tumorigenesis of colorectal cancer still partially understood. Therefore, this study was designed investigate significance MCAM cancer. expression analyzed by TCGA and GEO databases. qRT-PCR IHC analysis were conducted validate patient tissues. The tumor-inhibiting ability further assessed CCK-8 assay, colony formation wound-healing assay. WB...
To enable unmanned aerial vehicle (UAV) operators to efficiently and intuitively convey their commands a swarm of UAVs, we propose the use natural human-centric input modalities, such as voices gestures. This paper addresses fusion modalities voice gesture data, which are captured through microphone Leap Motion controller, respectively, control UAV swarms. The obtained experimental results presented, achieved performance (accuracy) is analyzed. Finally, combined human factor ergonomics test...
In this paper, a compact wideband dual-decoupling technique is proposed for microstrip patch antennas (WMPAs) at the X band. The decoupling property achieved by integrating defected ground structure (DGS) and coplanar inductance on plane layer, respectively. To validate feasibility of method, both aperture-coupled probe-fed WMPAs with novel scheme are simulated, fabricated, measured, An equivalent circuit characteristic mode analysis (CMA) developed to provide physical insight succinctly....
Segmenting lung fields from CT (Computed Tomography) scans is an important task for the analysis, diagnosis and treatment of pulmonary diseases. Although many segmentation methods have been presented, some new automatic l
The commonly used Deep Q Networks is known to overestimate action values under certain conditions.It's also proved that overestimations do harm performance, which might cause instability and divergence of learning.In this paper, we present the Sarsa (DSQN) algorithm, can considered as an enhancement algorithm.First, DSQN algorithm takes advantage experience replay target network techniques in improve stability neural networks.Second, double estimator utilized for Qlearning reduce...
Conventional methods that perform lung segmentation in CT rely on a large contrast Hounsfield units between the and surrounding tissues. But of lungs affected by high density pathologies are connected to border discontinuities pixel intensities may be caused X-ray projecting intensity changes, differing tissue reflectance transmission properties, so usual can't get good results. Here we proposed novel more effective algorithm used for segmenting fields images. The is based active contour...
Multi-agent deep reinforcement learning (MDRL) is an emerging research hotspot and application direction in the field of machine artificial intelligence. MDRL covers many algorithms, rules frameworks, it currently researched swarm system, energy allocation optimization, stocking analysis, sequential social dilemma, with extremely bright future. In this paper, a parallel-critic method based on classic algorithm MADDPG proposed to alleviate training instability problem cooperative-competitive...
Background: Aortic dissection refers to the separation of aortic media and extension along long axis form true false chambers wall. 65-70% patients died cardiac tamponade, arrhythmia, rupture, etc. At present, echocardiography, computed tomography angiography (CTA), are main diagnosis tools for dissection. To date, there is no rapid serum molecular marker that can be used differential risk assessment. Objectives: screen markers systematically amid acute coronary syndrome preliminarily...
In this paper, a low-cost compact decoupled bandwidth enhancement technique is proposed to address the high mutual coupling issue between wideband closely-spaced multiple-input and multiple-out (MIMO) patch antennas. The decoupling property realized by combining two circuits introduced defected ground structure (DGS) coplanar structure, respectively. An equivalent circuit given clarify mechanism provide physical insight. To validate feasibility of design scheme, prototypes arrays based on...
Nowadays, big data analytics has been widely applied in addressing the growing cybercrime threats. However, energy consumption is explosive increasing with fast growth of processing anti-cybercrime. In this paper, an energy-efficient framework for applications proposed to reduce while satisfying deadline constrains. First, problem tasks scheduling a single Spark job modeled as integer program. We design algorithm minimize application Spark. To avoid service-level agreement violations...