Using sEMG Signal Frequency to Evaluate Post-Stroke Elbow Spasticity.

Spasticity is a motor disorder with high prevalence and critical consequences following a stroke. Reliable and sensitive measurements are important to guide the selection and evaluation of treatment strategies. Technology-assisted methods, such as the surface electromyography (sEMG) technique, have been developed to measure spasticity as sensitive and accurate alternatives to commonly used clinical scales. However, sEMG amplitude based measures may confound spasticity-induced muscle activities with other types of muscle contractions. This study thus introduces the idea of using sEMG frequency information to detect spasticity as a potential solution to overcome the limitations of existing sEMG based measures. The preliminary results of three patients demonstrate the possibility and future research directions for this approach.

Experimental validation on dual-frequency outperforms single-frequency SSVEP with large numbers of targets within a given frequency range.

Multi-frequency steady-state visual evoked potential (SSVEP) aims to increase the number of targets in SSVEP-based brain-computer interfaces. However, the effectiveness of multi-frequency SSVEP when there is a large number of targets compared to traditional single-frequency SSVEP has not been demonstrated to date. It is also unclear the degree to which multi-frequency SSVEP outperforms single-frequency SSVEP as the number of targets increases. This study directly compares single-frequency and dual-frequency SSVEPs for different numbers of targets within a fixed (5 Hz) frequency range. Our results demonstrate that dual-frequency SSVEP maintains its performance at a high level of accuracy in the range while single-frequency SSVEP performance falls as the number of targets becomes very high within the given frequency range. In this particular study, dual-frequency SSVEP has a clear advantage when there are more than 120 targets in a 5 Hz frequency range.

Feasibility Evaluation of Online Classification-based Control for Gross Movement in a 2-DoF Prosthetic Arm.

Regression and classification models have been extensively studied to exploit the myoelectric and kinematic input information from the residual limb for the control of multiple degree-of-freedom (DoF) powered prostheses. The gross movement control of above-elbow prostheses is mainly based on regression models which map the available inputs to continuous prosthetic poses. However, the regression output is sensitive to the variation in the input signal. The myoelectric signal variation is usually large due to unintentional muscle contractions, which can deteriorate the user-in-the-loop performance with respect to the offline analysis. Alternatively, the classification models offer the advantage of being more robust to the input signal variation, but they were predominantly used for fine motor functions such as grasping. For gross motor functions, the discrete output may cause issues. Therefore, this work attempts to investigate the feasibility of utilising the classification model to control a 2-DoF transhumeral prosthesis for gross movement. The performance of 6 able-bodied subjects was evaluated in performing reaching and orientation matching tasks with a prosthetic arm in a virtual reality environment. The results were compared with the case of using their intact arms and existing results using the regression model. Our findings indicate that the classification-based method provides comparable performance to the regression model, making it a potential alternative for gross arm movement in multi-DoF prosthetic arms.

Differentiation in additive and multiplicative inputs to motoneuron pool as origins of spasticity - a neuromorphic modeling study.

Spasticity is characterized by a velocity-dependent increase in the tonic stretch reflex. Evidence suggests that spasticity originates from hyperactivity in the descending tract or reflex loop. To pinpoint the source of hyperactivity, however, is difficult due to lack of human data in-vivo. Thus, we implemented a neuromorphic model to revive the neurodynamics with spiking neuronal activity. Two types of input were modeled: (1) the additive condition (ADD) to apply tonic synaptic inputs directly into the reflex loop; (2) the multiplicative (MUL) condition to adjust the loop gains within the reflex loop. Results show that both conditions produced antagonist EMG responses resembling patient data. The timing of spasticity is more sensitive to the ADD condition, whereas the amplitude of spastic EMG is more sensitive to the MUL condition. In conclusion, our model shows that both additive and multiplicative hyperactivities suffice to elicit velocity-dependent spastic electromyographic signals (EMG), but with different sensitivities. This simulation study suggests that spasticity caused by different origins may be discernable by the progression of severity, which may help individualized goalsetting and parameter-selection in rehabilitation.Clinical Relevance-Potential application of neuromorphic modeling on spasticity includes selection of parameters for therapeutic plans, such as movement range, repetition, and load.

Modelling Physical Human-Robot Interface with Different Users, Cuffs, and Strapping Pressures: A Case Study.

Assisting persons during physical therapy or augmenting their performance often requires precise delivery of an intervention. Robotic devices are perfectly placed to do so, but their intervention highly depends on the physical human-robot connection. The inherent compliance in the connection leads to delays and losses in bi-directional power transmission and can lead to human-robot joint axes misalignment. This is often neglected in the literature by assuming a rigid connection and has a negative impact on the intervention's effectiveness and robustness. This paper presents the preliminary results of a study that aims to close that gap. The study investigates what model forms and parameters best capture human-robot connection dynamics across different persons, connection designs (cuffs), and cuff strapping pressures. The results show that the linear spring-damper model is the best compromise, but its parameters must be adjusted for each individual and different conditions separately.

A bioinspired 3D-printable flexure joint with cellular mechanical metamaterial architecture for soft robotic hands.

Compliant flexure joints have been widely used for cable-driven soft robotic hands and grippers due to their safe interaction with humans and objects. This paper presents a soft and compliant revolute flexure joint based on the auxetic cellular mechanical metamaterials with a heterogeneous structure. The heterogeneous architecture of the proposed metamaterial flexure joint (MFJ), which is inspired by the human finger joints, provides mechanically tunable multi-stiffness bending motion and large range of bending angle in comparison to conventional flexure joints. The multi-level variation of the joint stiffness over the range of bending motion can be tuned through the geometrical parameters of the cellular mechanical metamaterial unit cells. The proposed flexure joints are 3D printed with single flexible material in monolithic fashion using a standard benchtop 3D printer. The application of the MFJ is demonstrated in robotic in-hand manipulation and grasping thin and deformable objects such as wires and cables. The results show the capability and advantages of the proposed MFJ in soft robotic grippers and highly functional bionic hands.

Temporal and spatial characteristics of bone conduction as non-invasive haptic sensory feedback for upper-limb prosthesis.

Bone conduction is a promising haptic feedback modality for upper-limb prosthesis users, however, its potential and characteristics as a non-invasive feedback modality have not been thoroughly investigated. This study aimed to establish the temporal and spatial characteristics of non-invasive bone conduction as a sensory feedback interface for upper-limb prostheses. Psychometric human-subject experiments were conducted on three bony landmarks of the elbow, with a vibrotactile transducer affixed to each to provide the stimulus. The study characterized the temporal domain by testing perception threshold and resolution in amplitude and frequency. The spatial domain was evaluated by assessing the ability of subjects to detect the number of simultaneous active stimulation sites. The experiment was conducted with ten able-bodied subjects and compared to two subjects with trans-radial amputation. The psychometric evaluation of the proposed non-invasive bone conduction feedback showed results comparable to invasive methods. The experimental results demonstrated similar amplitude and frequency resolution of the interface for all three stimulation sites for both able-bodied subjects and subjects with trans-radial amputation, highlighting its potential as a non-invasive feedback modality for upper-limb prostheses.

Sensor Selection with Composite Features in Identifying User-Intended Poses for Human-Prosthetic Interfaces.

A Human-Prosthetic Interface (HPI) serves to estimate and realise the limb pose intended by the human user, using the information obtained from sensors worn by the user. In recent studies, the HPI maps multi-joint limb poses (i.e. coordinated movement of the body and limbs) to the inputs of multiple sensors. This is in contrast to the conventional methods where each degree of freedom of the powered prosthesis is mapped to the input of one/a pair of sensors. In this approach, it is necessary to systematically select sensors that carry the most information for the intended set of poses, to improve system accuracy and/or minimise the number of sensors, thus the complexity, in the prosthetic system. In this paper, sensor selection process is systematically formulated to maximise the information contained in the input features for a given number of sensors. Most importantly, it accounts for composite features, which are features requiring information from multiple sensors. Such composite features exist and are important in HPIs as we seek to capture coordinated motion involving movements of multiple limb and body segments. A non-convex optimisation problem is formulated which accounts for the constraint introduced by the composite features. A projection matrix is utilised as the optimisation variable to select intended features for evaluation. The problem is solved by the proposed Sensor Selection with Composite Features (SS-CF) algorithm which adapts convex-relaxation techniques. The SS-CF is benchmarked against HPI with expert-selected sensors in the literature and against a greedy heuristic method. The outcome demonstrated the efficacy of the SS-CF algorithm.

Polymeric fiber sensors for insertion forces and trajectory determination of cochlear implants in hearing preservation.

The level of hearing restoration in patients with severe to profound sensorineural hearing loss by means of cochlear implants (CIs) has drastically risen since the introduction of these neuroprosthetics. The proposed CI integrated with polymer optical fiber Bragg gratings (POFBGs) enables real-time evaluation of insertion forces and trajectory determination during implantation irrespective of the speed of insertion, as well as provides high signal quality, low stiffness levels, minimum induced stress even under forces of high magnitudes and exhibits significant reduction of the risk of fiber breakage inside the constricted cochlear geometry. As such, the proposed device opens new avenues towards atraumatic cochlear implantations and provides a direct route for the next generation of CIs with intraoperative insertion force assessment and path planning capacity crucial for surgical navigation. Hence, adaptation of this technology to clinical reality holds promising prospects for the hearing impaired.

Weak Monotonicity With Trend Analysis for Unsupervised Feature Evaluation.

Performance in an engineering system tends to degrade over time due to a variety of wearing or ageing processes. In supervisory controlled processes there are typically many signals being monitored that may help to characterize performance degradation. It is preferred to select the least amount of information to obtain high quality of predictive analysis from a large amount of collected data, in which labeling the data is not always feasible. To this end a novel unsupervised feature selection method, robust with respect to significant measurement disturbances, is proposed using the notion of "weak monotonicity" (WM). The robustness of this notion makes it very attractive to identify the common trend in the presence of measurement noises and population variation from the collected data. Based on WM, a novel suitability indicator is proposed to evaluate the performance of each feature. This new indicator is then used to select the key features that contribute to the WM of a family of processes when noises and variations among processes exist. In order to evaluate the performance of the proposed framework of the WM and suitability, a comparative study with other nine state-of-the-arts unsupervised feature evaluation and selection methods is carried out on well-known benchmark datasets. The results show a promising performance of the proposed framework on unsupervised feature evaluation in the presence of measurement noises and population variations.

Feature Identification Framework for Back Injury Risk in Repetitive Work With Application in Sheep Shearing.

BACKGROUND: Lower back injuries are a serious global problem. Most of these injuries occur over time with repeated sub-acute stresses. Neuromuscular control dysfunction could predict injury, however injuries are almost never observed alongside this data. No labels are available to identify important features that may be predictive of injury. While there are many individual differences in injury development, the population trend is that each individual's injury tolerance decreases over time with exposure, indicating a monotonic process. METHODS: This paper proposes a framework for identifying key features of injury using an unsupervised technique that exploits knowledge of injury aetiology by analysing which features contribute to the popular trend using weak monotonicity from data segmented by task repetitions. The feature selection also evaluates feature redundancy. The efficacy of the framework is demonstrated through data from on-site sheep shearers over one day using 17 wearable inertial measurement units and 16 surface electromyography (sEMG) sensors. RESULTS: Consistent with literature, the results demonstrate sEMG features derived from the erector spinae and multifidus muscles are the most important indicators for lower back injury. To evaluate the performance of the proposed population-trend based unsupervised feature selection technique, the self-reported fatigue information is treated as some 'ground truth' information so that this proposed technique can compare with 5 existing unsupervised feature selection techniques. CONCLUSION: The proposed technique is shown to be the most consistent with the self-reported fatigue information, demonstrating the effectiveness of the proposed method.

Technology-assisted assessment of spasticity: a systematic review.

BACKGROUND: Spasticity is defined as "a motor disorder characterised by a velocity dependent increase in tonic stretch reflexes (muscle tone) with exaggerated tendon jerks". It is a highly prevalent condition following stroke and other neurological conditions. Clinical assessment of spasticity relies predominantly on manual, non-instrumented, clinical scales. Technology based solutions have been developed in the last decades to offer more specific, sensitive and accurate alternatives but no consensus exists on these different approaches. METHOD: A systematic review of literature of technology-based methods aiming at the assessment of spasticity was performed. The approaches taken in the studies were classified based on the method used as well as their outcome measures. The psychometric properties and usability of the methods and outcome measures reported were evaluated. RESULTS: 124 studies were included in the analysis. 78 different outcome measures were identified, among which seven were used in more than 10 different studies each. The different methods rely on a wide range of different equipment (from robotic systems to simple goniometers) affecting their cost and usability. Studies equivalently applied to the lower and upper limbs (48% and 52%, respectively). A majority of studies applied to a stroke population (N = 79). More than half the papers did not report thoroughly the psychometric properties of the measures. Analysis identified that only 54 studies used measures specific to spasticity. Repeatability and discriminant validity were found to be of good quality in respectively 25 and 42 studies but were most often not evaluated (N = 95 and N = 78). Clinical validity was commonly assessed only against clinical scales (N = 33). Sensitivity of the measure was assessed in only three studies. CONCLUSION: The development of a large diversity of assessment approaches appears to be done at the expense of their careful evaluation. Still, among the well validated approaches, the ones based on manual stretching and measuring a muscle activity reaction and the ones leveraging controlled stretches while isolating the stretch-reflex torque component appear as the two promising practical alternatives to clinical scales. These methods should be further evaluated, including on their sensitivity, to fully inform on their potential.

Does Real-Time Feedback Improve User Performance in SSVEP-based Brain-Computer Interfaces?

Offline and online experiments are both widely used in SSVEP-based BCI research and development for different purposes. One of the major differences between offline and online experiments is the existence of real-time feedback to the user while they are using the interface. However, the role of feedback in SSVEP-based BCIs has not yet been well studied. This work focuses on understanding the effect of feedback in SSVEP-based BCIs and if there exists any relationship between offline and online BCI performance. An experiment was designed to compare directly the accuracies of the BCI with and without feedback for participants. Results showed that feedback can improve performance in a complex task, but no clear improvement was observed in a simple task.

A Practical Post-Stroke Elbow Spasticity Assessment Using an Upper Limb Rehabilitation Robot: A Validation Study.

Spasticity is a motor disorder characterised by a velocity-dependent increase in muscle tone, which is critical in neurorehabilitation given its high prevalence and potential negative influence among the post-stroke population. Accurate measurement of spasticity is important as it guides the strategy of spasticity treatment and evaluates the effectiveness of spasticity management. However, spasticity is commonly measured using clinical scales which may lack objectivity and reliability. Although many technology-assisted measures have been developed, showing their potential as accurate and reliable alternatives to standard clinical scales, they have not been widely adopted in clinical practice due to their low usability and feasibility. This paper thus introduces an easy-to-use robotic based measure of elbow spasticity and its evaluation protocol. Preliminary results collected with one post-stroke patient and one healthy control subject are presented and demonstrate the feasibility of the approach.

Exploring the Utility of Crutch Force Sensors to Predict User Intent in Assistive Lower Limb Exoskeletons.

The adoption of assistive lower limb exoskeletons in built environments is reliant on the further development of these devices to handle the varied conditions experienced in everyday life. The required development includes more varied and flexible gait patterns, but also appropriate user interfaces to enable fluid gait. This work explores the properties of an algorithm used to predict user intent based on sensors onboard a user-balanced robotic exoskeleton system. Specifically, classification algorithms built with different input data sets are compared - with varying detail of the interaction forces between the crutches and the ground, and the duration of the data sample used to make the prediction. Data were collected with one able-bodied participant using an exoskeleton, training three independent classifiers corresponding to different exoskeleton states. The results indicate the value of including information about the interaction forces between the crutches and the ground in improving prediction accuracy, with increasing prediction window also generally resulting in an increase in prediction accuracy. Whilst no categorical recommendation can be made with respect to either parameter, these results provide a baseline which can be used in conjunction deliberate consideration of the costs associated with implementation.

Investigating User Volitional Influence on Step Length in Powered Exoskeleton Designed for Users with SCI.

Volitional movement from users of assistive lower limb exoskeletons may be exploited to increase the controlled variability in the movements of a human-exoskeleton system. This may in turn allow these devices to handle the variability encountered in the terrain of everyday life. This study aimed to investigate the degree to which users can volitionally influence step length, when using an assistive exoskeleton designed for users with spinal cord injury (SCI) running a fixed robotic exoskeleton trajectory. An experiment was conducted to investigate the accessible range of step lengths when five able-bodied participants and one participant with SCI piloted a user-balanced exoskeleton. Participants were asked to take steps as large as possible ("large") and as small as possible ("small"), with the able-bodied individuals asked to minimise use of their leg muscles, with step length of each step measured. Surface electromyography (sEMG) data were collected on major leg muscles of the able-bodied subjects to monitor their muscle activities with a novel processing method introduced to facilitate discussion in the context of users with SCI. The results demonstrate that a user can intentionally manipulate the resulting step length, with every participant having significantly different large and small step sizes (p < 0.05). However, large variations were observed between individuals in terms of absolute step lengths and difference between large and small steps. Moreover, the range of step length (normalised by the leg length) ranged from 0.237 to 0.375 for the able-bodied subjects and 0.245 for the individual with SCI. Although positive correlation was present between the sEMG data and resulting step lengths, the result was not statistically significant (p > 0.05).

Varying Joint Patterns and Compensatory Strategies Can Lead to the Same Functional Gait Outcomes: A Case Study.

This paper analyses joint-space walking mechanisms and redundancies in delivering functional gait outcomes. Multiple biomechanical measures are analysed for two healthy male adults who participated in a multi-factorial study and walked during three sessions. Both participants employed varying intra- and inter-personal compensatory strategies (e.g., vaulting, hip hiking) across walking conditions and exhibited notable gait pattern alterations while keeping task-space (functional) gait parameters invariant. They also preferred various levels of asymmetric step length but kept their symmetric step time consistent and cadence-invariant during free walking. The results demonstrate the importance of an individualised approach and the need for a paradigm shift from functional (task-space) to joint-space gait analysis in attending to (a)typical gaits and delivering human-centred human-robot interaction.

Application of the extended technology acceptance model to explore clinician likelihood to use robotics in rehabilitation.

PURPOSE: Evidence suggests that patients with upper limb impairment following a stroke do not receive recommended amounts of motor practice. Robotics provide a potential solution to address this gap, but clinical adoption is low. The aim of this study was to utilize the technology acceptance model as a framework to identify factors influencing clinician adoption of robotic devices into practice. MATERIALS AND METHOD: Mixed methods including survey data and focus group discussions with allied health clinicians whose primary caseload was rehabilitation of the neurologically impaired upper limb. Surveys based on the technology acceptance measure were completed pre/post exposure to and use of a robotic device. Focus groups discussions based on the theory of planned behaviour were conducted at the conclusion of the study. RESULTS: A total of 34 rehabilitation clinicians completed the surveys with pre-implementation data indicating that rehabilitation clinicians perceive robotic devices as complex to use, which influenced intention to use such devices in practice. The focus groups found that lack of experience and time to learn influenced confidence to implement robotic devices into practice. CONCLUSION: This study found that perceived usefulness and perceived ease of use of a robotic device in clinical rehabilitation can be improved through experience, training and embedded technological support. However, training and embedded support are not routinely offered, suggesting there is a discordance between current implementation and the learning needs of rehabilitation clinicians.IMPLICATIONS FOR REHABILITATIONPatients do not receive adequate amounts of upper limb motor practice following a stroke, and although robotic devices have the potential to address this gap, clinical adoption is low.The technology acceptance model identified that clinicians perceive robotic devices to be complex to use with current implementation efforts failing to consider their training needs.Implementation adoption of robotic devices in rehabilitation should be supported with adequate training and technological support if sustainable practice change is to be achieved.

Promoting clinical best practice in a user-centred design study of an upper limb rehabilitation robot.

PURPOSE: Despite their promise to increase therapy intensity in neurorehabilitation, robotic devices have not yet seen mainstream adoption. Whilst there are a number of contributing factors, it is obvious that the treating clinician should have a clear understanding of the objectives and limitations of robotic device use. This study sought to explore how devices can be developed to support a clinician in providing clinical best practice. METHODS AND MATERIALS: A user-centred design study of a robotic device was conducted, involving build-then-use iterations, where successive iterations are built based on feedback from the use cycle. This work reports results of an analysis of qualitative and quantitative data describing the use of the robotic device in the clinical sessions, and from a focus group with the treating clinicians. RESULTS AND CONCLUSIONS: The data indicated that use of the device did not result in patient goal-setting and may have resulted in poor movement quality. Therapists expected a higher level of autonomy from the robotic device, and this may have contributed to the above problems. These problems can and should be addressed through modification of both the study design and device to provide more explicit instructions to promote clinical best practice.IMPLICATIONS FOR REHABILITATIONEncouraging clinical best practice when using evaluating prototype devices within a clinical setting is important to ensure that best practice is maintained - and can be achieved through both study and device designSupport from device developers can significantly improve the confidence of therapists during the use of that device in rehabilitation, particularly with new or prototype devicesEnd effector-based robotic devices for rehabilitation show potential for a wide variety of patient presentations and capabilities.

Frequency set selection for multi-frequency steady-state visual evoked potential-based brain-computer interfaces.

Objective: Multi-frequency steady-state visual evoked potential (SSVEP) stimulation and decoding methods enable the representation of a large number of visual targets in brain-computer interfaces (BCIs). However, unlike traditional single-frequency SSVEP, multi-frequency SSVEP is not yet widely used. One of the key reasons is that the redundancy in the input options requires an additional selection process to define an effective set of frequencies for the interface. This study investigates systematic frequency set selection methods. Methods: An optimization strategy based on the analysis of the frequency components in the resulting multi-frequency SSVEP is proposed, investigated and compared to existing methods, which are constructed based on the analysis of the stimulation (input) signals. We hypothesized that minimizing the occurrence of common sums in the multi-frequency SSVEP improves the performance of the interface, and that selection by pairs further increases the accuracy compared to selection by frequencies. An experiment with 12 participants was conducted to validate the hypotheses. Results: Our results demonstrated a statistically significant improvement in decoding accuracy with the proposed optimization strategy based on multi-frequency SSVEP features compared to conventional techniques. Both hypotheses were validated by the experiments. Conclusion: Performing selection by pairs and minimizing the number of common sums in selection by pairs are effective ways to select suitable frequency sets that improve multi-frequency SSVEP-based BCI accuracies. Significance: This study provides guidance on frequency set selection in multi-frequency SSVEP. The proposed method in this study shows significant improvement in BCI performance (decoding accuracy) compared to existing methods in the literature.