A5101393230- Topic Modeling
- Natural Language Processing Techniques
- Software Engineering Research
- Facial Trauma and Fracture Management
- Bone Tissue Engineering Materials
- Ocular Disorders and Treatments
- Electrospun Nanofibers in Biomedical Applications
- Wound Healing and Treatments
- 3D Printing in Biomedical Research
- Diabetic Foot Ulcer Assessment and Management
- PARP inhibition in cancer therapy
- Heat Transfer and Optimization
- Scientific Computing and Data Management
- Turbomachinery Performance and Optimization
- Laser Applications in Dentistry and Medicine
- Vehicular Ad Hoc Networks (VANETs)
- Cardiovascular Health and Disease Prevention
- Robot Manipulation and Learning
- Advanced Fluorescence Microscopy Techniques
- Simulation and Modeling Applications
- Radiative Heat Transfer Studies
- Advanced Steganography and Watermarking Techniques
- Cryptography and Data Security
- Ovarian cancer diagnosis and treatment
- Nasal Surgery and Airway Studies
Nanchang University
2025
Second Affiliated Hospital of Nanchang University
2025
Southern Medical University
2023-2024
Nanfang Hospital
2023-2024
University of Baltimore
2024
Johns Hopkins University
2024
China Academy of Engineering Physics
2023-2024
Johns Hopkins Medicine
2024
Second Affiliated Hospital of Zhejiang University
2021-2024
China University of Geosciences
2023
Abstract Diabetic wounds pose a challenge to healing due increased bacterial susceptibility and poor vascularization. Effective requires simultaneous biofilm elimination angiogenesis stimulation. In this study, we incorporated polyaniline (PANI) S-Nitrosoglutathione (GSNO) into polyvinyl alcohol, chitosan, hydroxypropyltrimethyl ammonium chloride chitosan (PVA/CS/HTCC) matrix, creating versatile wound dressing membrane through electrospinning. The combines the advantages of photothermal...
Abstract Background Poly ADP‐ribose polymerase (PARP) inhibitors are approved for the treatment of some men with advanced prostate cancer. Rare but serious side effects include myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). The impact PARP on clonal hematopoiesis (CH), a potential precursor lesion associated MDS AML, is incompletely understood in We hypothesized that would increase CH prevalence abundance. Methods prospectively enrolled participants cancer treated...
Prosthetic eye is indispensable as filler after enucleation in patients with anophthalmia, whereas there are still many complications including postoperative infection and socket depression or extrusion during the conventional artificial material applications. Some Ca-silicate biomaterials showed superior bioactivity but their biological stability vivo limit biomedical application long-term permanent implants. Herein we aimed to understand physicochemical potential responses of zinc doping...
Losing an eye to various reasons at any age may have a major impact on one's self-image, except for requiring adapt monocular vision. The conventional metal oxide or polymer implants orbital reconstruction, involving the vulnerable postoperative administration (migration, exposure), are always challenged due deficiency of some critical efficacy attainability. While coralline hydroxyapatite ceramics offer alternatives, they come with insufficient infection prevention and tailor-made...
Purpose This paper aims to present an intelligent motion attitude control algorithm, which is used solve the poor precision problems of motion-manipulation and balance humanoid robots. Aiming at a few physical training samples low efficiency, this proposes offline pre-training controller using identification model as priori knowledge online in real environment. Design/methodology/approach The deep reinforcement learning (DRL) continuous state space applied robots robot constructed. Combined...
Adaboost has been verified as an effective algorithm that improves the performance of weak learning algorithms which are slightly better than random guessing. In this paper, we use based feed forward neural network to solve inverse kinematics 6 Degrees-of-Freedom (DOF) robot with offset-wrist. The aims overcome dilemma most robots offset wrists can only be solved by numerical methods in low efficiency. Experiments indicate a good both accuracy and stability reducing more 70% average error...
Cultivated land not only is the foundation and lifeline of food production, but also an important premise basis for ensuring national security. Accurate identification farmland parcel great significance in agricultural production. This study proposed a dual-stream-fusion network (DSFNet), using semantic segmentation boundary detection parallel, high-resolution satellite images. We module based on DeepLabV3+ structure DexiNed to extract richer features by fusing information. Then,...
摘要: 为了解决仿人机器人运动控制精度和运动稳定性差等问题,提出智能运动姿态控制算法。将连续动作和连续状态空间的深度强化学习应用于姿态控制,建立机器人运动智能姿态控制器。并针对物理样机训练样本少、效率低等问题,提出使用机器人辨识模型对姿态控制器进行离线的预训练,作为真实物理环境下继续学习提升的先验知识,提高了后期训练效率。将优化后的机器人姿态控制器用于机器人的运动控制中,分别和加入PID控制器、MPC控制器、以及PID+MPC控制器的机器人运动相比,在环境过渡步行试验中机器人上身俯仰姿态轨迹跟踪残差标准差分别减少60.97%,46.36%,23.98%,在平地障碍物步行试验中机器人上身俯仰姿态轨迹跟踪残差标准差分别减少60.38%,26.38%,9.52%。
Immunotherapy and its evaluation have shown great promise for cancer treatment. Here, a mouse subcutaneous transplantable tumor model was applied to testing therapeutic strategies. The treated by regulating anti-PD-L1, anti-CTLA-4, cisplatin their combined therapy. Biochemistry experiments found that after immunotherapy, mice more immune responses, which were manifested an increase in the content of growth factors activation T cells. Meanwhile, multimodal nonlinear optical microscopy imaging...
Submodule capacitors are the most critical components for safe operation of modular multilevel converters (MMCs). It is very challenging to accurately monitor capacitor states without interrupting system's normal operation. This paper develops a novel non-intrusive capacitance monitoring algorithm adding external sensors. The inherent voltage and current relationship MMCs analyzed, an image identification method proposed identify correct inserted time window submodules (SMs). distribution...
In the prompt-specific holistic score prediction task for Automatic Essay Scoring, general approaches include pre-trained neural model, coherence and hybrid model that incorporate syntactic features with model. this paper, we propose a novel approach to extract represent essay prompt-learning NSP shows match state-of-the-art AES achieves best performance long essays. We apply feature dense embedding augment BERT-based achieve methodology AES. addition, explore various ideas combine...
The Internet of Things (IoT) has experienced rapid growth in recent years, drawing significant attention from both academics and industry. As IoT gradually gains popularity the industrial market, internet things(IIoT) emerged as a necessary solution to meet demands times. limitations posed by small storage space certain devices, restricted network throughput coupled with timing reliability requirements, need for safety assurance IIoT, all present challenges. In particular, security threats...
Unified Job-Description Language on Chemical-Grid (UDLC) is a domain-specific language designed to write full-function chemical jobs quickly by providing common language. It frees chemists focus problem-solving rather than the details of various scripting languages (CSLs) and target runtime platform associated middleware. UDLC can directly invoke grid services without dealing with interaction grid, such as job submission, monitoring. Thus, it interesting meaningful for translate scripts in...