A5064938053- Autonomous Vehicle Technology and Safety
- Reinforcement Learning in Robotics
- Traffic control and management
- Spine and Intervertebral Disc Pathology
- Traffic Prediction and Management Techniques
- Signaling Pathways in Disease
- Advanced Battery Materials and Technologies
- Adversarial Robustness in Machine Learning
- Cardiovascular, Neuropeptides, and Oxidative Stress Research
- Cell Adhesion Molecules Research
- Apelin-related biomedical research
- MicroRNA in disease regulation
- Advanced biosensing and bioanalysis techniques
- Musculoskeletal pain and rehabilitation
- Coronary Interventions and Diagnostics
- Inflammatory mediators and NSAID effects
- Protease and Inhibitor Mechanisms
- Biosensors and Analytical Detection
- Advanced Battery Technologies Research
- Nuclear Receptors and Signaling
- Protein Kinase Regulation and GTPase Signaling
- Cancer-related molecular mechanisms research
- Spinal Hematomas and Complications
- Infective Endocarditis Diagnosis and Management
- Advancements in Battery Materials
ShangHai JiAi Genetics & IVF Institute
2024
Shanghai Artificial Intelligence Laboratory
2024
University of Edinburgh
2023
Nanjing Brain Hospital
2020-2021
Nanjing Medical University
2020-2021
Shaanxi Provincial People's Hospital
2015-2021
Korea Maritime and Ocean University
2021
Harbin Medical University
2014-2020
Third Affiliated Hospital of Harbin Medical University
2014-2020
Nanjing Chest Hospital
2020
Driving safely requires multiple capabilities from human and intelligent agents, such as the generalizability to unseen environments, safety awareness of surrounding traffic, decision-making in complex multi-agent settings. Despite great success Reinforcement Learning (RL), most RL research works investigate each capability separately due lack integrated environments. In this work, we develop a new driving simulation platform called MetaDrive support generalizable reinforcement learning...
Diverse and realistic traffic scenarios are crucial for evaluating the AI safety of autonomous driving systems in simulation. This work introduces a data-driven method called TrafficGen scenario generation. It learns from fragmented human data collected real world then generates scenarios. is an autoregressive neural generative model with encoder-decoder architecture. In each iteration, it first encodes current context attention mechanism decodes vehicle's initial state followed by...
In recent years, we have witnessed the compelling application of Internet Things (IoT) in our daily life, ranging from living to industrial production. On account computation and power constraints, IoT devices offload their tasks remote cloud services. However, long-distance transmission poses significant challenges for latency-sensitive businesses, such as autonomous driving control. As a remedy, mobile edge computing (MEC) is deployed at network reduce delay. With MEC joining in, how...
Learning from active human involvement enables the subject to actively intervene and demonstrate AI agent during training. The interaction corrective feedback brings safety alignment learning process. In this work, we propose a new reward-free method called Proxy Value Propagation for policy optimization. Our key insight is that proxy value function can be designed express intents, wherein state-action pairs in demonstration are labeled with high values, while those agents' actions...
We have previously reported that nuclear factor of activated T cells (NFATs) play an important role in the regulation vascular smooth muscle cell migration and proliferation by receptor tyrosine kinase G protein-coupled agonists, platelet-derived growth factor-BB thrombin, respectively. To understand NFATs disease, we now studied involvement these transcription factors neointima formation a rat carotid artery balloon injury model. The levels NFATc1 injured right common arteries were...
To evaluate whether electroacupuncture (EA) at ST36 can accelerate the recovery of gastrointestinal motility after colorectal surgery.Forty patients American Society Anesthesiologists physical status II and III undergoing elective open resection malignant tumours were included in this study. Using a sealed envelope method, randomly divided into two groups either receiving EA (EA group) or sham (SEA group). Data regarding bowel function (times to first sounds, passage flatus defaecation)...
BACKGROUND: Propofol is a common sedative-hypnotic drug traditionally used for inducing and maintaining general anesthesia. Recent studies have drawn attention to the nonanesthetic effects of propofol, but potential mechanism by which propofol suppresses non–small-cell lung cancer (NSCLC) progression has not been fully elucidated. METHODS: For in vitro experiments, we (0, 2, 5, 10 µg/mL) treat A549 cells 1, 4, 12 hours Cell Counting Kit-8 (CCK-8) detect proliferation. Apoptosis was measured...
Abstract This study was aimed at investigating whether Elabela (ELA) gene therapy can promote angiogenesis in the treatment of myocardial infarction (MI). The fusion expression plasmid pAAV‐3 × Flag/ELA‐32 successfully constructed using molecular cloning technique. model acute MI established by ligating left anterior descending coronary artery mice. Adeno‐associated virus serotype 9 (AAV9) injected into surrounding myocardium and tail vein immediately after established. AAV again from one...
Self-Driven Particles (SDP) describe a category of multi-agent systems common in everyday life, such as flocking birds and traffic flows. In SDP system, each agent pursues its own goal constantly changes cooperative or competitive behaviors with nearby agents. Manually designing the controllers for system is time-consuming, while resulting emergent are often not realistic nor generalizable. Thus simulation remains challenging. Reinforcement learning provides an appealing alternative...
Large-scale driving datasets such as Waymo Open Dataset and nuScenes substantially accelerate autonomous research, especially for perception tasks 3D detection trajectory forecasting. Since the logs in these contain HD maps detailed object annotations which accurately reflect real-world complexity of traffic behaviors, we can harvest a massive number complex scenarios recreate their digital twins simulation. Compared to hand-crafted often used existing simulators, data-driven collected from...
Interleukin (IL)-6 induced vascular smooth muscle cell (VSMC) motility in a dose-dependent manner. In addition, IL-6 stimulated tyrosine phosphorylation of gp130, resulting the recruitment and activation STAT-3. IL-6-induced VSMC was found to be dependent on gp130/STAT-3 signaling. also cyclin D1 expression time- gp130/STAT-3-dependent manner VSMCs. Suppression levels via use its small interfering RNA molecules inhibited motility. Furthermore, balloon injury both at mRNA protein rat carotid...
Platelet-derived growth factor BB (PDGF-BB) induced cyclin A expression and CDK2 activity in vascular smooth muscle cells (VSMC). Inhibition of nuclear factors activated T cell (NFAT) activation by cyclosporin (CsA) VIVIT suppressed PDGF-BB-induced activity, resulting blockade VSMC the G(1) phase. In addition, CsA- VIVIT-mediated inhibition NFATs small interfering RNA-targeted down-regulation levels DNA synthesis. PDGF-BB also mRNA an NFAT-dependent manner. Cloning bioinformatic analysis rat...
To understand the role of epidermal growth factor receptor (EGFR) transactivation in G protein-coupled (GPCR) agonist-induced signaling events, we have studied capacity thrombin activation Gab1-SHP2 vascular smooth muscle cells (VSMCs). Thrombin activated both Gab1 and SHP2 EGFR-dependent manner. Similarly, induced Rac1 Cdc42 activation, these responses were suppressed when either or stimulation is blocked. also PAK1 a time- EGFR-Gab1-SHP2-Rac1/Cdc42-dependent Inhibition EGFR, Gab1, SHP2,...
Abstract Apela was recently identified as a new ligand of the apelin peptide jejunum (APJ) receptor. The purpose this study to investigate role apela in post‐myocardial infarction (post‐MI) recovery from cardiorenal damage. A murine MI model established, and then infused subcutaneously for two weeks. Echocardiographs were performed before after at indicated times. Renal function evaluated by serum urine biochemistry. Immunohistochemistry heart kidney tissue situ terminal deoxynucleotidyl...
The p73 protein, a paralogue of the p53 tumor suppressor, is essential for normal development and survival neurons. TP73 therefore interest as candidate gene Alzheimer's disease (AD) susceptibility. mRNA transcribed from three promoters, termed P1 – P3, there evidence an additional complexity in its regulation, namely, variable allelic expression bias some human tissues. We utilized RT-PCR/RFLP direct cDNA sequencing to measure allele-specific mRNA, SNP genotyping assess genetic associations...
Previously, we have demonstrated that STAT-3 plays a role in thrombin-induced VSMC motility. To learn more about the of STATs mitogenic and chemotactic signaling events thrombin, here studied STAT-5. Thrombin activated STAT-5 as measured by its tyrosine phosphorylation, DNA binding, reporter gene activity. Inhibition STAT-5B, but not STAT-5A, adenovirus-mediated expression respective dominant-negative mutants suppressed growth induced Hsp27 FGF-2 time- STAT-5B-dependent manner VSMC. In...