An EMG-controlled neuroprosthesis for daily upper limb support: a preliminary study

Implementation of a Surface Electromyography-Based Upper Extremity Exoskeleton Controller Using Learning from Demonstration

Upper-extremity exoskeletons have demonstrated potential as augmentative, assistive, and rehabilitative devices. Typical control of upper-extremity exoskeletons have relied on switches, force/torque sensors, and surface electromyography (sEMG), but these systems are usually reactionary, and/or rely on entirely hand-tuned parameters. sEMG-based systems may be able to provide anticipatory control, since they interface directly with muscle signals, but typically require expert placement of sensors on muscle bodies. We present an implementation of an adaptive sEMG-based exoskeleton controller that learns a mapping between muscle activation and the desired system state during interaction with a user, generating a personalized sEMG feature classifier to allow for anticipatory control. This system is robust to novice placement of sEMG sensors, as well as subdermal muscle shifts. We validate this method with 18 subjects using a thumb exoskeleton to complete a book-placement task. This learning-from-demonstration system for exoskeleton control allows for very short training times, as well as the potential for improvement in intent recognition over time, and adaptation to physiological changes in the user, such as those due to fatigue.

Classification of Anticipatory Signals for Grasp and Release from Surface Electromyography

Surface electromyography (sEMG) is a technique for recording natural muscle activation signals, which can serve as control inputs for exoskeletons and prosthetic devices. Previous experiments have incorporated these signals using both classical and pattern-recognition control methods in order to actuate such devices. We used the results of an experiment incorporating grasp and release actions with object contact to develop an intent-recognition system based on Gaussian mixture models (GMM) and continuous-emission hidden Markov models (HMM) of sEMG data. We tested this system with data collected from 16 individuals using a forearm band with distributed sEMG sensors. The data contain trials with shifted band alignments to assess robustness to sensor placement. This study evaluated and found that pattern-recognition-based methods could classify transient anticipatory sEMG signals in the presence of shifted sensor placement and object contact. With the best-performing classifier, the effect of label lengths in the training data was also examined. A mean classification accuracy of 75.96% was achieved through a unigram HMM method with five mixture components. Classification accuracy on different sub-movements was found to be limited by the length of the shortest sub-movement, which means that shorter sub-movements within dynamic sequences require larger training sets to be classified correctly. This classification of user intent is a potential control mechanism for a dynamic grasping task involving user contact with external objects and noise. Further work is required to test its performance as part of an exoskeleton controller, which involves contact with actuated external surfaces.

MUNDUS project: MUltimodal neuroprosthesis for daily upper limb support

Background

MUNDUS is an assistive framework for recovering direct interaction capability of severely motor impaired people based on arm reaching and hand functions. It aims at achieving personalization, modularity and maximization of the user’s direct involvement in assistive systems. To this, MUNDUS exploits any residual control of the end-user and can be adapted to the level of severity or to the progression of the disease allowing the user to voluntarily interact with the environment. MUNDUS target pathologies are high-level spinal cord injury (SCI) and neurodegenerative and genetic neuromuscular diseases, such as amyotrophic lateral sclerosis, Friedreich ataxia, and multiple sclerosis (MS). The system can be alternatively driven by residual voluntary muscular activation, head/eye motion, and brain signals. MUNDUS modularly combines an antigravity lightweight and non-cumbersome exoskeleton, closed-loop controlled Neuromuscular Electrical Stimulation for arm and hand motion, and potentially a motorized hand orthosis, for grasping interactive objects.

Methods

The definition of the requirements and of the interaction tasks were designed by a focus group with experts and a questionnaire with 36 potential end-users.

Five end-users (3 SCI and 2 MS) tested the system in the configuration suitable to their specific level of impairment. They performed two exemplary tasks: reaching different points in the working volume and drinking. Three experts evaluated over a 3-level score (from 0, unsuccessful, to 2, completely functional) the execution of each assisted sub-action.

Results

The functionality of all modules has been successfully demonstrated. User’s intention was detected with a 100% success. Averaging all subjects and tasks, the minimum evaluation score obtained was 1.13 ± 0.99 for the release of the handle during the drinking task, whilst all the other sub-actions achieved a mean value above 1.6. All users, but one, subjectively perceived the usefulness of the assistance and could easily control the system. Donning time ranged from 6 to 65 minutes, scaled on the configuration complexity.

Conclusions

The MUNDUS platform provides functional assistance to daily life activities; the modules integration depends on the user’s need, the functionality of the system have been demonstrated for all the possible configurations, and preliminary assessment of usability and acceptance is promising.

First study towards linear control of an upper-limb neuroprosthesis with an EEG-based Brain-Computer Interface

In this study we show how healthy subjects are able to use a non-invasive Motor Imagery (MI)-based Brain Computer Interface (BCI) to achieve linear control of an upper-limb neuromuscular electrical stimulation (NMES) controlled neuroprosthesis in a simple binary target selection task. Linear BCI control can be achieved if two motor imagery classes can be discriminated with a reliability over 80% in single trial. The results presented in this work show that there was no significant loss of performance using the neuroprosthesis in comparison to MI where no stimulation was present. However, it is remarkable how different the experience of the users was in the same experiment. The stimulation either provoked a positive reinforcement feedback, or prevented the user from concentrating in the task.