Fei Wang

ORCID: 0009-0001-8341-3497
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • EEG and Brain-Computer Interfaces
  • Robot Manipulation and Learning
  • Gaze Tracking and Assistive Technology
  • Sleep and Work-Related Fatigue
  • Prosthetics and Rehabilitation Robotics
  • Muscle activation and electromyography studies
  • Sleep and related disorders
  • Soft Robotics and Applications
  • Neuroscience and Neural Engineering
  • Robotic Locomotion and Control
  • Heart Rate Variability and Autonomic Control
  • Advanced Surface Polishing Techniques
  • Blind Source Separation Techniques
  • Hand Gesture Recognition Systems
  • Tactile and Sensory Interactions
  • Robotics and Sensor-Based Localization
  • Stroke Rehabilitation and Recovery
  • Video Surveillance and Tracking Methods
  • Inertial Sensor and Navigation
  • Markov Chains and Monte Carlo Methods
  • Target Tracking and Data Fusion in Sensor Networks
  • Emotion and Mood Recognition
  • Sleep and Wakefulness Research
  • Traditional Chinese Medicine Analysis
  • ECG Monitoring and Analysis

Jiangsu University of Science and Technology
2024-2025

Northeastern University
2014-2024

Zhejiang Sci-Tech University
2024

Huawei Technologies (United Kingdom)
2023-2024

Shaoxing University
2024

Ningbo Center for Disease Control and Prevention
2023

Ningbo University Affiliated Hospital
2023

Wuhan City Chinese Medicine Hospital
2023

Ningbo University
2023

China Academy of Launch Vehicle Technology
2023

Gynostemma pentaphyllum (GP) is widely used for the treatment of diseases such as hyperlipidemia, fatty liver and obesity in China, atorvastatin broadly an anti-hyperlipidemia drug. This research focuses on plasma metabolites following four groups rats: control, a hyperlipidemia model, model treated with GP atorvastatin. Using (1)H-NMR-based metabonomics, we elucidated therapeutic mechanisms Orthogonal Partial Least Squares-Discriminant analysis (OPLS-DA) plotting metabolic state potential...

10.1371/journal.pone.0078731 article EN cc-by PLoS ONE 2013-11-01

The mental state of a driver can be accurately and reliably evaluated by detecting the driver’s electroencephalogram (EEG) signals. However, traditional machine learning deep methods focus on single electrode feature analysis ignore functional connection brain. In addition, recent brain function network method needs to manually extract substantial features, which results in cumbersome operation. For this reason, paper introduces graph convolution combined with theory into study fatigue...

10.1063/5.0008434 article EN Review of Scientific Instruments 2020-07-01

Robot grasping has become a very hot research field so that the requirements for robot operation are getting higher and higher. In previous studies, use of traditional target detection algorithms is often inefficient, this article dedicated to improving deep reinforcement learning algorithm improve efficiency solve problem robots dealing with impact unknown disturbances on grasping. Using characteristic actively explores environment, Gaussian parameter Deep Deterministic Policy Gradient...

10.1063/5.0034101 article EN Review of Scientific Instruments 2021-02-01

Fatigue is a state commonly caused by overworked, which seriously affects daily work and life. How to detect mental fatigue has always been hot spot for researchers explore. Electroencephalogram (EEG) considered one of the most accurate objective indicators. This article investigated development classification algorithms applied in EEG-based detection recent years. According different source data, we can divide these into two categories, intra-subject (within same subject) cross-subject...

10.32604/iasc.2023.029698 article EN cc-by Intelligent Automation & Soft Computing 2022-08-18

Human–Robot Interaction In article number 2200050, Geng Yang and co-workers implement the co-design of deep learning algorithms tactile sensor. By utilizing neural networks (DNNs), sensor can effectively detect location magnitude external force recognize different touch modalities. Besides, a novel data augmentation method is developed based on sensor's rotation symmetry structure, which enhances DNNs' generalization performance.

10.1002/aisy.202270027 article EN cc-by-nc Advanced Intelligent Systems 2022-06-01

<sec> <title>BACKGROUND</title> The current landscape of Emergency Care (EC) is marked by high demand leading to issues such as Department boarding, overcrowding and subsequent delays that impact the quality safety patient care. Integrating data science into EC can enhance decision-making with predictive, preventative, personalized, participatory approaches. However, gaps in adherence fairness, accountability, interpretability, responsibility are evident, particularly due barriers...

10.2196/preprints.74202 preprint EN cc-by 2025-03-22

<sec> <title>BACKGROUND</title> Emergency departments (EDs) face significant challenges due to overcrowding, prolonged waiting times, and staff shortages, leading increased strain on health care systems. Efficient triage systems accurate departmental guidance are critical for alleviating these pressures. Recent advancements in large language models (LLMs), such as ChatGPT, offer potential solutions improving patient outpatient department selection emergency settings. </sec>...

10.2196/preprints.71613 preprint EN 2025-01-22

Based on a small, lightweight, low-cost high performance inertial Measurement Units(IMU), an effective calibration method is implemented to evaluate the of Micro-Electro-Mechanical Systems(MEMS) sensors suffering from various errors get acceptable navigation results.A prototype development board based FPGA, dual core processor's configuration for INS/GPS integrated system designed experimental testing.The significant error sources IMU such as bias, scale factor, and misalignment are...

10.22266/ijies2011.0630.04 article EN International journal of intelligent engineering and systems 2011-06-30

Abstract Objective. Parkinson’s disease (PD) is one of the most common neurodegenerative diseases, and early diagnosis crucial to delay progression. The PD has always been a difficult clinical problem due lack reliable biomarkers. Electroencephalogram (EEG) detection method, studies have attempted discover EEG spectrum characteristics PD, but reported conclusions are not uniform heterogeneity patients. There an urgent need for more advanced algorithm extract from satisfy personalized...

10.1088/1741-2552/ac40a0 article EN Journal of Neural Engineering 2021-12-01

Gynostemma pentaphyllum (GP) is widely used for the treatment of diseases such as hyperlipidemia, fatty liver and obesity in China, atorvastatin a popular anti-hyperlipidemia drug. Based on GC/MS metabonomic methods, this study focuses comparative metabolic profiling analysis acids plasma hyperlipidemic rats before after treatment, order to identify mechanism two drugs. Results obtained show that GP can effectively improve levels saturated acids, although has only relatively weak regulatory...

10.1039/c4ay01405g article EN Analytical Methods 2014-08-26

In recent years, Brain Computer Interface (BCI) based on motor imagery has been widely used in the fields of medicine, active safe systems for automobiles, entertainment, and so on. Motor relevant electroencephalogram (EEG) signals are weak, nonlinear, susceptible to interference. As a feature extraction method imagery, Common Spatial Pattern (CSP) proven be very effective. However, its effectiveness depends heavily choice frequency bands, Euclidean space cannot effectively describe inner...

10.1063/1.5142343 article EN Review of Scientific Instruments 2020-03-01

Two complexity parameters of EEG, i.e. sample entropy and rhythm energy are utilized to characterize the irregularity EEG data under different mental fatigue states. Then wavelet transform BP neural networks combined differentiate two The WT is employed extract nonlinear features from improve generalization performance BPNN. investigation suggests that can effectively describe dynamic which strongly correlated with fatigue. Both significantly decreased as level increases. These may be used...

10.1109/cyber.2015.7288238 article EN 2015-06-01

Motor preparation and execution require the interactions of a large-scale brain network, while study dynamic changes their could uncover underlying neural mechanism corresponding information processing. This analysis requires high temporal resolution recorded signals. Electroencephalogram (EEG) with has been widely used in related studies. However, studies based on scalp EEG always lead to distorted results, due volume conduction, compared that cortically In current study, networks motor are...

10.1109/tnsre.2016.2608359 article EN IEEE Transactions on Neural Systems and Rehabilitation Engineering 2016-01-01

With the rapid development of urbanization and motorization in China, fatigue driving has become an increasingly serious road traffic problem. Driving affects drivers' alertness, decreasing individual's ability to operate a vehicle safely increasing risk human error that could lead fatalities, which have been widely recognized as critical safety issues cut across all modes transportation industry. In this paper, firstly, with virtual system we developed, simulation experiments were designed...

10.1109/icra.2014.6907445 article EN 2014-05-01

Introduction The exact relationship between long-term shift work (SW) and cognitive impairment (CI) has been poorly understood. effects of the rotating night SW (RNSW) combining daytime recharge (DTR) on function were investigated. Methods A total 920 retired nurses 656 female teachers aged ≥50 years analyzed. Participants who worked at least once per week for 8 hat more than 1 year defined as group, those without a regular nighttime control group. associations among duration, frequency, DTR...

10.3389/fnagi.2021.827772 article EN cc-by Frontiers in Aging Neuroscience 2022-01-25

Abstract Robotics has been booming in recent years. Especially with the development of artificial intelligence, more and researchers have devoted themselves to field robotics, but there are still many shortcomings multi-task operation robots. Reinforcement learning achieved good performance manipulator manipulation, especially grasping, grasping is only first step for robot perform actions, it often ignores stacking, assembly, placement, other tasks be carried out later. Such long-horizon...

10.1007/s10846-024-02078-3 article EN cc-by Journal of Intelligent & Robotic Systems 2024-04-01
Coming Soon ...