The ISEK Congress will include a series of workshops included in your registration fee and scheduled to take place on Sunday, July 12, 2020. Please see the various descriptions below.
Workshop 1 – Multidisciplinary approaches to quantify sensorimotor control and adaptation of skilled and pathological hand movements
Presenters: Marco Santello, Arizona State University; Tomohiko Takei, Kyoto University; Shinichi Furuya, Sony Inc.; Natsue Yoshimura, Tokyo Institute of Technology
Neural control of the human hand relies on complex interactions among multiple levels of the nervous system. Traumatic and neurological injuries challenge the nervous system’s ability to control skilled hand movements. The goals of the workshop are (1) to provide an overview of the hand’s control system and biomechanics and (2) describe how complementary research approaches can advance our understanding of physiological sensorimotor control of the hand and provide insights into rehabilitation approaches for improving sensorimotor function. Dr. Tomohiko Takei’s presentation will focus on design of electrophysiological experiments to investigate neural mechanisms for sensorimotor control in non-human models (primates) and humans. Dr. Takei will describe techniques including single-unit recordings from cerebral cortex and spinal cord, as well as large-scale electromyography (EMG) and neuromodulatory approaches (Cooling and DREADD). Dr. Marco Santello’s presentation will address design of behavioral approaches for inferring sensorimotor control mechanisms (e.g., grasping and dexterous manipulation at unconstrained contacts; response inhibition) and integration with neuromodulation and electrophysiological approaches for system identification purposes. Dr. Santello will present approaches through published and ongoing studies consisting of electroencephalography (EEG), electromyography (EMG), single-pulse transcranial magnetic stimulation, continuous theta burst stimulation, and transcranial focused ultrasound stimulation for system identification purposes. Dr. Shinichi Furuya will describe design of data science approaches for extracting sensorimotor skills and disorders using multivariate analyses and machine learning techniques in highly trained individuals. Dr. Furuya will describe approaches including a combination of recording with motion capture and EMG with dimensionality reduction analyses (e.g. PCA and NMF) and penalized regression analyses (e.g. LASSO and ridge regression), which identify specific kinematic and neuromuscular features associated with enhancement and degradation of motor performance. Dr. Natsue Yoshimura will describe design of behavioral and neuroimaging approaches for estimating sensorimotor control and sensory feedback using non-invasive methods. Through the integration of robotic (KINARM) and machine learning approaches, Dr. Yoshimura will cover several data analytical methods, such as cortical current source estimation, neural decoding, EEG-based connectivity analysis, functional and structural MRI, and diffusion tensor images (DTI). The workshop takeaway knowledge or material that attendees will acquire include: basic understanding of experimental design, approaches, and analytical methods to identify neural control mechanisms and adaptation processes.
Workshop 2 – Advanced assessment of mechanical and neural properties of isometric and dynamic contractions: Perspectives for sports performance and sports medicine
Presenters: Anderson Oliveira, Aalborg University; Francesco Negro, University of Brescia; Eduardo Martinez-Valdes, University of Birmingham; Andrew Cresswell, University of Queensland; Alberto Botter, Politecnico de Torino
Assessing the pattern of muscle contraction is an essential part of research dedicated to describe or modify human movement. There are methodological advances towards improving the accuracy and relevance of data extracted during muscle actions, which are also highly relevant for professionals involved in sports performance and sports medicine worldwide. For instance, surface EMG has been extensively used to describe neural control of movements, despite widely known limitations regarding the extraction of neural strategies to control movements during dynamic EMG recordings. These limitations might be overcome using combined methods such as ultra-sound and high-density EMG, and professionals from sports and exercise sciences may highly benefit from such advanced assessments. The goals of this workshop are: 1) To provide basic background on the current methods to assess mechanical and neural properties of the muscle function and exploring their limitations in the assessment of isometric or dynamic contractions; 2) To provide an overview on the usage of both ultra-sound and multi-channel surface EMG for the assessment of muscle mechanics and neural modulation, specifically in exercise and sport sciences; 3) To demonstrate the strengths and limitations of using ultra-sound and multichannel surface EMG to describe muscle function in dynamic contractions, offering participants hands-on experiences on such method and subsequent data analysis.
Workshop 3 – High-definition transcranial direct current stimulation: Theory and practical considerations
Presenters: Shapour Jaberzadeh, Monash University; Maryam Zoghi, La Trobe University
Background: Neuronal electrophysiology (NE) is the study of the electrical properties of biological cells and tissues in the nervous system. NE is essential for understanding of brain function. Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation (NIBS) technique which can be used for modulation of NE. This modulation can be used as a probe for better understanding of the brain functions during control of human movements. It can be also used for enhancement of physiological processes such as motor performance, cognitive-sensory-motor control of human movements and physiological processes such skill aquisition and learning. Finaly, tDCS could be also used in rehabilation of patients with pathological conditions including movement/balance disorders, i.e. Parkinson disease and stroke. High-definition tDCS (HD-tDCS) is a novel technique for application of tDCS. It enable the users to target cortical and even deeper brain structures with weak direct currents. HD-tDCS is non-invasive, targeted, and can leverage the therapeutic potential, convenience, safety. <b>Aims:</b> The main goals of this workshop are: 1. to explain how HD-tDCS modulates neuronal electrophysiology. 2. to describe how this modulation affects sensory, motor and cognitive aspects of human movements. 3. to show how tDCS current modeling can be used for dosage determination 4. to enable the participants to understand the practical aspects of HD-tDCS application. The workshop promotes expertise in the field of NIBS and focuses on building competencies in the following areas: Principles and mechanisms of HD-tDCS, electrode preparation and montages, stimulation parameters, protocols and targeted outcomes, safety, HD-tDCS current modeling and dose determination, good practices in HD-tDCS delivery, HD-tDCS clinical trial design and implementation. Education and training will be provided by a highly experienced team of physiotherapists/neuroscientists. Previous practical experience with tDCS not required. The workshop utilises versatile educational elements from the basic to expert level and covers the complex framework from neurophysiological research to clinical applications. <b>Relevance:</b> This workshop provides an overview of the principles and state-of-the-art applications of HD-tDCS. Due to the multidisciplinary nature of ISEK and focus of the workshop on the technique for modulation of neural electrophysiology and its effects on human movements, this workshop is very relevant to the members of this event which include biomedical scientists, engineers, physical educationists, physical therapists and many other disciplines. Takeaway skill from this workshop include theoretical and practical aspects of HD-tDCS application for targeting and modulation of function in different brain areas. All participants will be provided workshop slide handouts, and a certificate.
Workshop 4 – Interpretable predictive modelling using functional physiological variables.
Presenters: Bernard Liew, University of Essex; David Rugamer, Ludwig-Maximilians-Universität München
Background: Predicting health outcomes, such as the probability of recovery is commonly undertaken by clinicians and movement scientists. A predictive model contains the best combination of covariates needed to achieve the best predictive accuracy. Covariates can come from various sources, such as demographic variables. A rich, yet underutilised source of covariates for prediction, is from bio-signals-such as muscle activation patterns. Incorporating bio-signals into predictive models can be challenging (e.g. large numbers of covariates exceeding sample size) which may deter its more widespread inclusion in predictive modelling. Bio-signals collected by movement scientist can be scalar (e.g. peak angle) and functional (e.g. angle waveform) in nature. Even though functional variables provide a richer mechanistic insight into an individual’s health than scalar variables, the former demand for special care and adequate generalizations of common statistical methods. Aim: The aim of this workshop is to equip audiences with sufficient knowledge and skills around FDboost, to perform classification using scalar and functional covariates. The workshop will first begin with a lecture by Dr Ruegamer, giving a basic introduction into functional data analysis, functional regression models and advantages using boosting. Dr Ruegamer is actively engaged in the development of the theory and software around FDboost, and his lecture will provide audiences with cutting edge knowledge of contemporary machine learning techniques. Next, Dr Liew will introduce our recent works using FDboost applied within the context of clinical biomechanics. Dr Liew, being a physiotherapist, is in an ideal position to inform audiences about how FDboost can solve real-world problems faced by clinicians and researchers. The workshop will transition to a hands-on session, where we will familiarise audiences to the open-source R software. Dr Ruegamer will demonstrate the pre-processing steps to prepare the data, analyse the data using FDboost, tune the model to achieve optimal performance, and validate the model. Finally, Dr Liew will illustrate how to interpret the model’s solutions using coefficient and partial dependency plots. The speakers will demonstrate these steps using previously published data. The attendees will than have the opportunity to apply FDboost to their dataset. Lastly, we provide a brief commentary on how to report the Methods and Results in a typical journal article, and leave room for a question and answer session. Relevance: There is a growing awareness that functional statistical techniques (e.g. Statistical Parametric Mapping) provide a richer understanding of human movement and its complexities. To our knowledge, all of the functional statistical technique used have focused on hypothesis testing. The proposed workshop provides an ideal opportunity to harness the information embedded within functional movement variables for clinical prediction.
Workshop 5 – Transferring the high-density surface EMG-based motor unit identification to practice: recent progresses and challenges
Presenters: Ales Holobar, University of Maribor; Deren Barsakcioglu, Imperial College London
Motor unit identification from high-density surface electromyograms (HDEMG) is maturing fast and gaining popularity in different research fields. Research groups around the globe are proving its potential in relatively large variety of neurophysiological investigations. However, its transfer to everyday neurologic, neurophysiologic, rehabilitation and ergonomic practice is still a challenge that needs full scientific and engineering attention. Quality control and the representativeness of the identified neural codes is still not fully established and while being non-invasive, the HDEMG acquisition is still not unobtrusive. Also the computational power needed for identification of motor units in real time calls for the novel technical solutions. In this workshop, we will present and discuss the recent progresses in the HDEMG acquisition and analysis. The focus will be on the quality control of motor unit identification and real-time neural codes identification, but the other classical topics of motor unit identification and neural information extraction in various experimental conditions will also be addressed and discussed. Different case studies will be presented, along with explanations of their main methodological steps.