A5025711005- Robotics and Sensor-Based Localization
- Energy Efficient Wireless Sensor Networks
- Distributed Control Multi-Agent Systems
- Advanced Vision and Imaging
- Soft Robotics and Applications
- Robotic Path Planning Algorithms
- Indoor and Outdoor Localization Technologies
- Modular Robots and Swarm Intelligence
- Robot Manipulation and Learning
- Wireless Body Area Networks
- Surgical Simulation and Training
- Control and Dynamics of Mobile Robots
- Micro and Nano Robotics
- Non-Invasive Vital Sign Monitoring
- ECG Monitoring and Analysis
- Minimally Invasive Surgical Techniques
- Robotic Locomotion and Control
- Underwater Vehicles and Communication Systems
- Context-Aware Activity Recognition Systems
- Inertial Sensor and Navigation
- Advanced Chemical Sensor Technologies
- Sparse and Compressive Sensing Techniques
- Augmented Reality Applications
- Medical Imaging and Analysis
- Advanced Image and Video Retrieval Techniques
University of Tennessee at Knoxville
2015-2024
Southwest Hospital
2009-2024
Army Medical University
2009-2024
Scunthorpe General Hospital
2024
Knoxville College
2017-2021
University of Toronto
2018-2021
Sichuan University
2021
Scuola Superiore Sant'Anna
2021
Lockheed Martin (United States)
2021
University of Hong Kong
2021
In this paper, a novel time division multiple access based MAC protocol designed for body sensor networks (BSNs) is presented. H-medium-access control (MAC) aims to improve BSNs energy efficiency by exploiting heartbeat rhythm information, instead of using periodic synchronization beacons, perform synchronization. Heartbeat inherent in every human and observable various biosignals. Biosensors BSN can extract the from their own sensory data detecting waveform peaks. All rhythms represented...
The design and structure of a self-assembly modular robot (Sambot) are presented in this paper. Each module has its own autonomous mobility can connect with other modules to form robotic structures different manipulation abilities. Sambot versatile, robust, flexible structure. computing platform provided for each is distributed consists number interlinked microcontrollers. interaction connectivity between achieved through infrared sensors Zigbee wireless communication discrete state control...
High-resolution, real-time data obtained by human motion tracking systems can be used for gait analysis, which helps better understanding the cause of many diseases more effective treatments, such as rehabilitation outpatients or recovery from lost motor functions after a stroke. In order to achieve ambulatory with low-cost MARG (magnetic, angular rate, and gravity) sensors, computationally efficient robust algorithm orientation estimation is critical. This paper presents an analytically...
Food recognition is a key component in evaluation of everyday food intakes, and its challenge due to intraclass variation. In this paper, we present an automatic classification method, DietCam, which specifically addresses the variation appearances. DietCam consists two major components, ingredient detection classification. ingredients are detected through combination deformable part-based model texture verification model. From ingredients, categories classified using multiview multikernel...
The relay node placement problem for wireless sensor networks is concerned with placing a minimum number of nodes into network to meet certain connectivity and survivability requirements. We study constrained versions the problem, where can only be placed at subset candidate locations. In connected we want place ensure base stations. survivable biconnectivity For each two problems, discuss its computational complexity, present framework polynomial time O(1) -approximation algorithms small...
This paper proposes the recognition framework of car makes and models from a single image captured by traffic camera. Due to various configurations cameras, may be in different viewpoints lighting conditions, quality varies resolution color depth. In framework, cars are first detected using part-based detector, license plates headlamps as cardinal anchor points rectify projective distortion. Car features extracted, normalized, classified an ensemble neural-network classifiers. experiment,...
To improve the efficiency of deep reinforcement learning (DRL)-based methods for robotic trajectory planning in unstructured working environment with obstacles. Different from traditional sparse reward function, this paper presents two brand-new dense functions. First, azimuth function is proposed to accelerate process locally a more reasonable by modeling position and orientation constraints, which can reduce blindness exploration dramatically. further efficiency, at subtask-level provide...
In this paper we present an approach to decoupled force/position control of the end-effector along same direction for redundant robots, and nonholonomic cart pushing with mobile manipulators. The manipulator is considered as a robot, unified dynamic model integrated platform on-board developed. linearized using nonlinear feedback technique in frame. Combining event-based planning method singularity analysis robot arm, task level action controller designed online kinematic redundancy...
This paper presents a hidden Markov model (HMM) approach for real-time activity classification using signals from wearable wireless sensor networks. A network can be used to continuously monitor the daily activities of subject in real time. However, nodes are constrained by limited battery and computing resources. The proposed HMM framework has been applied find most probable states series with low data transmission rate, which makes it highly suitable applications. performance was evaluated...
Obesity prevention requires individuals to have healthy eating and physical activity awareness in their daily lives. New technological products been explored help monitoring individual health behaviors. Smart phones are ubiq-uitous hand-held tools with rich embedded sensors limited computing power. A smart phone is a prospective candidate integrate exercise analysis food intake monitoring. This paper presents novel health-aware system (HealthAware) which utilizes the accelerometer monitor...
The purpose of this paper is to develop an automatic camera phone based multi-view food classifier as part a intake assessment system. Food important for obesity management, which has shown significant impacts in public healthcare. Conventional dietary record methods exhibit insufficient popularity due their low accuracy and high dependence on human interactions. Image recognition appears recently. But it still under development far away from field applications. This presents DietCam,...
This paper presents a novel visual-inertial integration system for human navigation in free-living environments, where the measurements from wearable inertial and monocular visual sensors are integrated. The preestimated orientation, obtained magnet, angular rate, gravity sensors, is used to estimate translation based on data sensors. has significant effect performance of fusion sensing strategy makes procedure much easier, because gravitational acceleration can be correctly removed...
In this paper, BSN-MAC, a medium access control (MAC) protocol designed for Body Sensor Networks (BSNs) is proposed. Due to the traffic coupling and sensor diversity characteristics of BSNs, common MAC protocols can not satisfy unique requirements biomedical sensors in BSNs. BSN-MAC exploits feedback information from deployed form closed-loop parameters. A algorithm proposed enable BSN coordinator adjust parameters IEEE 802.15.4 superframe achieve both energy efficiency low latency on...
H-MAC is a novel Time Division Multiple Access (TDMA) based MAC protocol designed for Body Sensor Networks (BSNs). It improves energy efficiency by exploiting human heartbeat rhythm information to perform time synchronization TDMA. Heartbeat inherent in every body and can be detected variety of biosignals. Therefore, biosensors BSNs extract the from their sensory data. Moreover, all rhythms represented peak sequences are naturally synchronized since they driven same source, heartbeat. By...
This paper presents a novel approach to ECG signal filtering and classification. Unlike the traditional techniques which aim at collecting processing signals with patient being still, lying in bed hospitals, our proposed algorithm is intentionally designed for monitoring classifying patient’s free-living environment. The patients are equipped wearable ambulatory devices whole day, facilitates real-time heart attack detection. In preprocessing, an integral-coefficient-band-stop (ICBS) filter...
Metabolites can double as a signaling modality that initiates physiological adaptations. Metabolism, chemical language encoding biological information, has been recognized powerful principle directing inflammatory responses. Cytosolic pH is regulator of response in macrophages. Here, we found L-malate exerts anti-inflammatory effect via BiP-IRF2BP2 signaling, which sensor cytosolic First, L-malate, TCA intermediate upregulated pro-inflammatory macrophages, was identified potent metabolite...
The goal of the present study was to investigate if cognitive traces network structure phonological language network, where word-form neighbors are connected each other, could be uncovered in word substitution errors. has a set macro-level (i.e., features characterizing global lexicon) and meso-level intermediate or subgroups within structural that should observable speech error data such play role production retrieval processes. 1,067 single errors, which included 965 errors slips tongue)...
Abstract Improved genetic characterization of rhabdomyosarcoma (RMS) subtypes, including PAX3/PAX7 fusion-negative RMS (FN-RMS) with RAS pathway mutations [1], has led to hope for individualized treatments fewer side effects than conventional chemotherapy. Understanding the metabolic vulnerabilities remains an area room improvement. We hypothesized that determining inhibition in RAS-active could inform new therapeutic combinations. performed polar metabolomic profiling comparing trametinib...