A5015752410- Muscle activation and electromyography studies
- Neuroscience and Neural Engineering
- EEG and Brain-Computer Interfaces
- Motor Control and Adaptation
- Computational Drug Discovery Methods
- Chemical Reactions and Isotopes
- Protein Structure and Dynamics
- Advanced Sensor and Energy Harvesting Materials
- Hand Gesture Recognition Systems
- Machine Learning in Materials Science
- Gaze Tracking and Assistive Technology
- vaccines and immunoinformatics approaches
- Neural Networks and Applications
- Telemedicine and Telehealth Implementation
- Advanced Memory and Neural Computing
- Hearing Loss and Rehabilitation
- Industrial Technology and Control Systems
- SARS-CoV-2 and COVID-19 Research
- Magnetic Field Sensors Techniques
- Speech and Audio Processing
- Service and Product Innovation
- Electrical and Bioimpedance Tomography
- Prosthetics and Rehabilitation Robotics
- Atomic and Subatomic Physics Research
- Blind Source Separation Techniques
Newcastle University
2018-2022
University of Edinburgh
2013-2019
University of Southampton
2013
Myoelectric pattern recognition systems can decode movement intention to drive upper-limb prostheses. Despite recent advances in academic research, the commercial adoption of such remains low. This limitation is mainly due lack classification robustness and a simultaneous requirement for large number electromyogram (EMG) electrodes. We propose address these two issues by using multi-modal approach which combines surface electromyography (sEMG) with inertial measurements (IMs) an appropriate...
In the field of upper-limb myoelectric prosthesis control, use statistical and machine learning methods has been long proposed as a means enabling intuitive grip selection actuation. Recently, this paradigm found its way toward commercial adoption. Machine learning-based control typically relies on large number electrodes. Here, we propose an end-to-end strategy for multi-grip, classification-based using only two sensors, comprising electromyography (EMG) electrodes inertial measurement...
The reconstruction of finger movement activity from surface electromyography (sEMG) has been proposed for the proportional and simultaneous myoelectric control multiple degrees-of-freedom (DOFs). In this paper, we propose a framework assessing decoding performance on novel movements, that is movements not included in training dataset. We then use our to compare linear kernel ridge regression sEMG accelerometry. Our findings provide evidence that, although non-linear method superior seen by...
Machine learning-based myoelectric control is regarded as an intuitive paradigm, because of the mapping it creates between muscle co-activation patterns and prosthesis movements that aims to simulate physiological pathways found in human arm. Despite that, there has been evidence closed-loop interaction with a classification-based interface results user adaptation, which leads performance improvement experience. Recently, focus shift toward continuous control, yet little known about whether...
Prosthetic devices for hand difference have advanced considerably in recent years, to the point where mechanical dexterity of a state-of-the-art prosthetic approaches that natural hand. Control options users, however, not kept pace, meaning new are used their full potential. Promising developments control technology reported literature met with limited commercial and clinical success. We previously described biomechanical model could be prosthesis control. The goal this study was evaluate...
The COVID-19 pandemic continues to pose a substantial threat human lives and is likely do so for years come. Despite the availability of vaccines, searching efficient small-molecule drugs that are widely available, including in low- middle-income countries, an ongoing challenge. In this work, we report results open science community effort, "Billion molecules against challenge", identify inhibitors SARS-CoV-2 or relevant receptors. Participating teams used wide variety computational methods...
Recent work on myoelectric prosthetic control has shown that the incorporation of accelerometry information along with surface electromyography (sEMG) potential improving performance and robustness a device by increasing classification accuracy. In this study, we investigated whether could further benefit from use additional sensory modalities such as gyroscopes magnetometers. We trained multi-class linear discriminant analysis (LDA) classifier to discriminate between six hand grip patterns...
Computational methods for virtual screening can dramatically accelerate early-stage drug discovery by identifying potential hits a specified target. Docking algorithms traditionally use physics-based simulations to address this challenge estimating the binding orientation of query protein–ligand pair and corresponding affinity score. Over recent years, classical modern machine learning architectures have shown outperforming traditional docking algorithms. Nevertheless, most learning-based...
Magnetomyography utilizes magnetic sensors to record small fields produced by the electrical activity of muscles, which also gives rise electromyogram (EMG) signal typically recorded with surface electrodes. Detection and recording these requires sensitive possibly equipped a CMOS readout system. This paper presents highly Hall sensor fabricated in standard $0.18~\mu \mathrm {m}$ technology for future low-field MMG applications. Compared previous works, our experimental results show that...
People who either use an upper limb prosthesis and/or have used services provided by a prosthetic rehabilitation centre, hereafter called users, are yet to benefit from the fast-paced growth in academic knowledge within field of prosthetics. Crucially over past decade, research has acknowledged limitations conducting laboratory-based studies for clinical translation. This led increase, albeit rather small, trials that gather real-world user data. Multi-stakeholder collaboration is critical...
The ultimate goal of machine learning-based myoelectric control is simultaneous and independent multiple degrees freedom (DOFs), including wrist digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Unfortunately, such have thus far met with limited success. In this work, we propose action decoding, a paradigm-shifting approach for independent, multi-digit...
One way of enhancing the dexterity powered myoelectric prostheses is via proportional and simultaneous control multiple degrees-of-freedom (DOFs). Recently, it has been demonstrated that reconstruction finger movement feasible by using features surface electromyogram (sEMG) signal. In such paradigms, number predictors target variables usually large, strong correlations are present in both input output domains. Synergistic patterns sEMG space have previously exploited to facilitate kinematics...
People who either use an upper limb prosthesis and/or have used services provided by a prosthetic rehabilitation centre, experience limitations of currently available devices. Collaboration between academia and broad range stakeholders, can lead to the development solutions that address peoples' needs. By doing so, rate device abandonment decrease. Co-creation is approach enable collaboration this nature occur throughout research process. We present findings co-creation project gained user...
Improving the robustness of myoelectric control to work over many months without need for recalibration could reduce prosthesis abandonment. Current approaches rely on post-hoc error detection verify certainty a decoder's prediction using predefined threshold value. Since decoder is fixed, performance decline time inevitable. Other such as supervised and unsupervised self-recalibration entail limitations in scaling up computational resources. The objective this paper study active learning...
Linear discriminant analysis (LDA) is the most commonly used classification method for movement intention decoding from myoelectric signals. In this work, we review performance of various variants on task hand motion classification. We demonstrate that optimal achieved with regularized (RDA), a which generalizes class-conditional Gaussian classifiers, including LDA, quadratic (QDA), and naive Bayes (GNB). The RDA offers continuum between these models via tuning two hyper-parameters control...
Recent studies indicate the limited clinical acceptance of myoelectric prostheses, as upper extremity amputees need improved functionality and more intuitive, effective, coordinated control their artificial limbs. Rather than exclusively classifying electromyogram (EMG) signals, it has been shown that inertial measurements (IMs) can form an excellent complementary signal to EMG signals improve prosthetic robustness. We present investigation into possibility replacing, rather complementing,...
Motor cortical local field potentials (LFPs) have been successfully used to decode both kinematics and kinetics of arm movement. For future clinically viable prostheses, however, brain activity decoders will generalize well under a wide spectrum behavioral conditions. This property has not yet demonstrated clearly. Here, we provide evidence for the first time, that an LFP-based electromyogram (EMG) decoder can reasonably across two different types behavior. We implanted intracortical...
Cochlear implants (CIs) require efficient speech processing to maximize information transmission the brain, especially in noise. A novel CI strategy was proposed our previous studies, which sparsity-constrained non-negative matrix factorization (NMF) applied envelope order improve performance noisy environments. It showed that algorithm needs be adaptive, rather than fixed, adjust acoustical conditions and individual characteristics. Here, we explore benefit of a system allows user signal...
Modern, commercially available hand prostheses offer the potential of individual digit control. However, this feature is often not utilized due to lack a robust scheme for finger motion estimation from surface electromyographic (EMG) measurements. Regression methods have been proposed achieve closed-loop position, velocity, or force In paper, we propose an alternative approach, based on open-loop action-based control, which could be achieved through simultaneous classification. We compare...
Abstract In the field of upper-limb myoelectric prosthesis control, use statistical and machine learning methods has been long proposed as a means enabling intuitive grip selection actuation. Recently, this paradigm found its way toward commercial adoption. Machine learning-based control typically relies on large number electrodes. Here, we propose an end-to-end strategy for multi-grip, classification-based using only two sensors, comprising electromyography (EMG) electrodes inertial...
Abstract Objective We aim to develop a paradigm for simultaneous and independent control of multiple degrees freedom (DOFs) upper-limb prostheses. Approach introduce action , novel method operate prosthetic digits with surface electromyography (EMG) based on multi-label, multi-class classification. At each time step, the decoder classifies movement intent controllable DOF into one three categories: open, close, or stall (i.e., no movement). implemented real-time myoelectric system using this...
Abstract Prosthetic devices for hand difference have advanced considerably in recent years, to the point where mechanical dexterity of a state-of-the-art prosthetic approaches that natural hand. Control options users, however, not kept pace, meaning new are used their full potential. Promising developments control technology reported literature met with limited commercial and clinical success. We previously described biomechanical model could be prosthesis control. In this study, we report...
ABSTRACT The ultimate goal of machine learning-based myoelectric control is simultaneous and independent multiple degrees freedom (DOFs), including wrist digit artificial joints. For prosthetic finger control, regression-based methods are typically used to reconstruct position/velocity trajectories from surface electromyogram (EMG) signals. Although such have produced highly-accurate results in offline analyses, their success real-time prosthesis settings has been rather limited. In this...