A5057528184- Spine and Intervertebral Disc Pathology
- Gait Recognition and Analysis
- Indoor and Outdoor Localization Technologies
- Diabetic Foot Ulcer Assessment and Management
- Medical Imaging and Analysis
- Modular Robots and Swarm Intelligence
- Industrial Vision Systems and Defect Detection
- Reinforcement Learning in Robotics
- Robotic Path Planning Algorithms
- Emotion and Mood Recognition
- Balance, Gait, and Falls Prevention
- Bone fractures and treatments
- Color perception and design
- Evolutionary Algorithms and Applications
- Face and Expression Recognition
- Domain Adaptation and Few-Shot Learning
- Wireless Networks and Protocols
- Remote-Sensing Image Classification
- Water Quality Monitoring Technologies
- Inertial Sensor and Navigation
- Elbow and Forearm Trauma Treatment
- Advanced Optical Network Technologies
- Antiplatelet Therapy and Cardiovascular Diseases
- Neural Networks and Applications
- Advanced Neural Network Applications
Hefei University of Technology
2023-2025
Hunan University
2014-2025
University of Electronic Science and Technology of China
2020-2024
First Affiliated Hospital of Dalian Medical University
2010-2022
Dalian Medical University
2010-2022
Shandong University of Science and Technology
2022
Alibaba Group (China)
2021
Xuzhou Medical College
2017-2018
Shanghai Institutes for Biological Sciences
2018
Jimei University
2014
Foot-mounted inertial navigation is an important issue in areas such as pedestrian localization, gait analysis, and sport training. However, low-cost sensors suffer from several errors that make the results less convincing. In this paper, a multi-sensor approach with one sensor on each foot presented to reduce system heading drift. Through dual-gait parameters between two feet are employed non-collocated uncoupled subsystems be related other. A step length estimator based inverted pendulum...
Gait analysis is a prosperous tool for the clinical evaluation and diagnosis. In this article, portable gait system based on foot-mounted inertial sensors established. A threshold-free method using long short-term memory recurrent neural network constructed to segment four typical phases in sequence temporal parameters analysis. Segmentation accuracy reaches over 95% across recruited subjects with distinct patterns, which significantly superior when compared traditional machine learning...
This paper presents a versatile multi-sensor fusion method and decision-making algorithm for ambulatory continuous patient monitoring purposes via body sensor network (BSN). Gait features including spatio-temporal parameters, gait asymmetry, regularity were identified estimated from individual patients data collected clinical trials. Hence, assessment diagnosis of the improvement or deterioration lower limb rehabilitation process is ensured. The experimental results 10-m free walking trials...
Device-free localization based on WiFi Channel State Information (CSI) has attracted considerable attention in recent years. Many of the schemes are learning-based, among which CSI fingerprinting deep learning is most promising. Learning-based assume distributions fingerprints keep relatively stable over time. However, this assumption often does not hold. As signals prone to be influenced by various environmental factors, vary significantly with changes. Consequently, model built for...
Platycodin D (PD) is one of the major bioactive components roots Platycodon grandiflorum and possesses multiple biological pharmacological properties, such as antiviral, anti-inflammatory, anti-cancer activities. However, whether it affects platelet function remains unclear. This study aims to evaluate role PD in thrombus formation. Platelets were treated with followed by measuring aggregation, activation, spreading, clot retraction, expression glycoprotein receptors. Moreover, mice...
The measured noise impedance of MOSFETs is found to be invariant across technology nodes. Together with the invariance optimum figure current density, JOPT, this allows for optimally matched LNAs ported without redesign between nodes and a given design scaled in frequency. porting frequency scaling are validated experimentally on record low fabricated 90nm 130nm CMOS technology.
Document clustering is one of the key problems in text mining and information retrieval area. It groups documents a way that maximizes similarity within clusters minimizes between different clusters. Most partitioning based algorithms are sensitive to initial centroids, result greatly depends on centroids. This paper first uses unsupervised feature selection method reduce dimension document space then proposes novel algorithm which select cluster centriods process by size density datasets....
In this paper, from the perspective of human ergonomics, we analyze movement joints in process body movements, and establish a dynamic model according to skeleton structure. On basis, rigid dynamics point view, combined with principle
Coating defect detection is a critical aspect of ensuring product quality in the manufacturing process. However, due to variety coating defects and complex background actual production, detecting these can be challenging. To improve accuracy robustness detection, method based on data augmentation network optimization design proposed. First, feature image random adaptive weighted mapping (FIRAWM) strategy proposed, considering prior accuracy, quantity, context information each category. Then,...
XCS [1], the accuracy-based classifier system, provides a very brilliant way to merge genetic algorithmic (GA) rule learning and reinforcement (RL) methodologies together. This makes it suitable for wide range of applications where generalisation over decision making states is desirable. Also, its Q-learning-oriented prediction update scheme enables handle multi-step problems adequately. paper reports how intertwined spirals problem, initially popular benchmark in classification, was...
The development of artificial intelligence (AI) technologies has significantly improved the performance automatic facial expression recognition. Expression recognition technology is applied to lots fields such as mood and fatigue detection. In this paper, a thermal comfort detection method developed based on frown adjusted network structure presented. To explore relationship between expressions thermal, we also design experiments collect data by setting temperature only environmental...
Research projects about evolution of agents in a cellular world are not new topics the artificial life (AL) fields. However, most studies focus on those fundamental, social behaviours like energy preservation, pattern formation or leader following etc. This paper presents experiments applications genetic algorithms (GAs) to an empirical multiple robot cooperative task: unknown environment exploration. These investigate effectiveness GAs for evolving individual swarm members that constitute...
Designing control processes for mobile robot teams is a very difficult task. Consequently, methods the to learn or adapt on their own are highly desirable. Genetic algorithms (GAs), which class of techniques inspired by biological evolution, mainstream method accomplish this self-adaptation. However, major problem genetic algorithm approach brittleness evolved solution(s). The solutions provided GAs can only work properly in those environments closely resembling training environment. A...
Objective: To summarize the clinical data of elastic stable intramedullary nailing(ESIN) in treatment long bone fracture children a single medical center, and to analyze problems occurred after ESIN surgery corresponding solutions. Methods: A retrospective analysis was conducted regarding 2 133 pediatric fractures conforming inclusion exclusion criteria from June, 2005 December, 2017 Department Orthopedics, Children's Hospital Nanjing Medical University.There were 1 191 boys 942 girls, aged...
For hyperspectral data classification, feature reduction techniques have become an apparent need to extract information from original data. In this paper, we introduce Locality Sensitive Discriminant Analysis (LSDA) perform for classification of imagery. By preserving both the discriminant and local geometrical structure in data, proposed method can obtain better accuracy with less computational load. Experiments carried out on two sets, Indian Pine Pavia City, show that results our...
Learning temporal dependencies among targets (TDT) benefits better time series forecasting, where refer to the predicted sequence. Although autoregressive methods model TDT recursively, they suffer from inefficient inference and error accumulation. We argue that integrating learning into non-autoregressive is essential for pursuing effective efficient forecasting. In this study, we introduce differencing approach represent propose a parameter-free plug-and-play solution through an...