A5040842669- Muscle activation and electromyography studies
- Machine Learning and ELM
- Neural Networks and Applications
- Face and Expression Recognition
- Prosthetics and Rehabilitation Robotics
- EEG and Brain-Computer Interfaces
- Motor Control and Adaptation
- Balance, Gait, and Falls Prevention
- Advanced Sensor and Energy Harvesting Materials
- Gaze Tracking and Assistive Technology
- Stroke Rehabilitation and Recovery
- Neuroscience and Neural Engineering
- Non-Invasive Vital Sign Monitoring
- Extracellular vesicles in disease
- Hand Gesture Recognition Systems
- Cerebral Palsy and Movement Disorders
- MicroRNA in disease regulation
- Vestibular and auditory disorders
Imperial College London
2022-2025
UK Dementia Research Institute
2024
Nankai University
2023
NIHR Imperial Biomedical Research Centre
2023
China Agricultural University
2018-2020
Abstract Globally, the prevalence of stroke is significant and increasing annually. This growth has led to a demand for rehabilitation services that far exceeds supply, leaving many survivors without adequate rehabilitative care. In response this challenge, study introduces portable exoskeleton system integrates neural control mechanisms governing human arm movements. design leverages neuroplasticity principles simulate natural movements, aiming reactivate strengthen neuromuscular...
To control wearable robotic systems, it is critical to obtain a prediction of the user's motion intent with high accuracy. Surface electromyography (sEMG) recordings have often been used as inputs for these devices, however bipolar sEMG electrodes are highly sensitive their location. Positional shifts after training gait models can therefore result in severe performance degradation. This study uses high-density (HD-sEMG) simulate various electrode signals from four leg muscles during...
The neural control of human quiet stance remains controversial, with classic views suggesting a limited role the brain and recent findings conversely indicating direct cortical muscles during upright posture. Conceptual feedback models have been proposed tested against experimental evidence. most renowned model is continuous impedance model. However, when time delays are included in this to simulate transmission, controller becomes unstable. Another model, intermittent assumes that central...
Prediction of transition between locomotion modes (e.g. moving from flat ground to stairs, etc) is vital for optimal interface with lower limb assistive technologies such as exoskeletons and prostheses. Inertial bipolar electromyography (EMG) sensors have been investigated, but accuracy clinical utility remains unresolved. This shortfall may be attributed their limited capacity detect subtle changes in muscle activations, particularly during the early stages transitions (e.g., near toe-off)....
Existing methods for human locomotion mode recognition often rely on using multiple bipolar electrode sensors muscle groups to accurately identify underlying motor activities. To avoid this complex setup and facilitate the translation of technology, we introduce a single grid high-density surface electromyography (HDsEMG) electrodes mounted location (above rectus femoris) classify six modes in walking. By employing neural network, trained model achieved average accuracy 97.7% with 160ms...
<p>For the control of wearable robotics, it is critical to obtain a prediction user’s motion intent with high accuracy. Electromyography (EMG) recordings have often been used as inputs for these devices, however bipolar EMG electrodes are highly sensitive their location. Positional shifts after training gait models can therefore result in severe performance degradation. </p> <p>This study uses high-density simulate various electrode signals from four leg muscles during...
<p>For the control of wearable robotics, it is critical to obtain a prediction user’s motion intent with high accuracy. Electromyography (EMG) recordings have often been used as inputs for these devices, however bipolar EMG electrodes are highly sensitive their location. Positional shifts after training gait models can therefore result in severe performance degradation. </p> <p>This study uses high-density simulate various electrode signals from four leg muscles during...
Applying semi-supervised learning to extreme machine (ELM), we propose a classification framework (SSELM) with arbitrary norm (q-norm, q=0,1 and 2). However, the SSELM involves nonconvex nonsmooth problem. In this work, two types of optimizatio n methods are developed solve proposed SSELM. The first one is an exact solution approach that reformulates as mixed integer programming. second approximation approximates by DC (difference convex functions) Several formulations for presented...