A Comparative Analysis of Speed Profile Models for Ankle Pointing Movements: Evidence that Lower and Upper Extremity Discrete Movements are Controlled by a Single Invariant Strategy

Statistical evaluation of tongue capability with visual feedback

BACKGROUND

Analysis of tongue movement would benefit from a reference showcasing healthy tongue capability. We aimed to develop a reference of tongue capability and evaluated the role of visual feedback on the expression of movement.

METHODS

Using a wireless tracking intraoral wearable device, we composed probability distributions of the tongue tip as subjects were asked to explore the entire sensing surface area. Half of the 32 subjects received live visual feedback of the location of the center of the tongue tip contact.

RESULTS

We observed that the visual feedback group was 51.0% more consistent with each other in the position domain, explored 21.5% more sensing surface area, and was 50.7% more uniformly distributed. We found less consistent results when we evaluated velocity and acceleration.

CONCLUSION

Visual feedback best established a healthy capability reference which can be used for designing new interfaces, quantifying tongue ability, developing new diagnostic and rehabilitation techniques, and studying underlying mechanisms of tongue motor control.

A neurophysiologically interpretable deep neural network predicts complex movement components from brain activity

The effective decoding of movement from non-invasive electroencephalography (EEG) is essential for informing several therapeutic interventions, from neurorehabilitation robots to neural prosthetics. Deep neural networks are most suitable for decoding real-time data but their use in EEG is hindered by the gross classes of motor tasks in the currently available datasets, which are solvable even with network architectures that do not require specialized design considerations. Moreover, the weak association with the underlying neurophysiology limits the generalizability of modern networks for EEG inference. Here, we present a neurophysiologically interpretable 3-dimensional convolutional neural network (3D-CNN) that captured the spatiotemporal dependencies in brain areas that get co-activated during movement. The 3D-CNN received topography-preserving EEG inputs, and predicted complex components of hand movements performed on a plane using a back-drivable rehabilitation robot, namely (a) the reaction time (RT) for responding to stimulus (slow or fast), (b) the mode of movement (active or passive, depending on whether there was an assistive force provided by the apparatus), and (c) the orthogonal directions of the movement (left, right, up, or down). We validated the 3D-CNN on a new dataset that we acquired from an in-house motor experiment, where it achieved average leave-one-subject-out test accuracies of 79.81%, 81.23%, and 82.00% for RT, active vs. passive, and direction classifications, respectively. Our proposed method outperformed the modern 2D-CNN architecture by a range of 1.1% to 6.74% depending on the classification task. Further, we identified the EEG sensors and time segments crucial to the classification decisions of the network, which aligned well with the current neurophysiological knowledge on brain activity in motor planning and execution tasks. Our results demonstrate the importance of biological relevance in networks for an accurate decoding of EEG, suggesting that the real-time classification of other complex brain activities may now be within our reach.

Restoration of bilateral motor coordination from preserved agonist-antagonist coupling in amputation musculature

Background

Neuroprosthetic devices controlled by persons with standard limb amputation often lack the dexterity of the physiological limb due to limitations of both the user’s ability to output accurate control signals and the control system’s ability to formulate dynamic trajectories from those signals. To restore full limb functionality to persons with amputation, it is necessary to first deduce and quantify the motor performance of the missing limbs, then meet these performance requirements through direct, volitional control of neuroprosthetic devices.

Methods

We develop a neuromuscular modeling and optimization paradigm for the agonist-antagonist myoneural interface, a novel tissue architecture and neural interface for the control of myoelectric prostheses, that enables it to generate virtual joint trajectories coordinated with an intact biological joint at full physiologically-relevant movement bandwidth. In this investigation, a baseline of performance is first established in a population of non-amputee control subjects (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n = 8$$\end{document}n=8). Then, a neuromuscular modeling and optimization technique is advanced that allows unilateral AMI amputation subjects (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n = 5$$\end{document}n=5) and standard amputation subjects (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n = 4$$\end{document}n=4) to generate virtual subtalar prosthetic joint kinematics using measured surface electromyography (sEMG) signals generated by musculature within the affected leg residuum.

Results

Using their optimized neuromuscular subtalar models under blindfolded conditions with only proprioceptive feedback, AMI amputation subjects demonstrate bilateral subtalar coordination accuracy not significantly different from that of the non-amputee control group (Kolmogorov-Smirnov test, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P \ge 0.052$$\end{document}P≥0.052) while standard amputation subjects demonstrate significantly poorer performance (Kolmogorov-Smirnov test, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P < 0.001$$\end{document}P<0.001).

Conclusions

These results suggest that the absence of an intact biological joint does not necessarily remove the ability to produce neurophysical signals with sufficient information to reconstruct physiological movements. Further, the seamless manner in which virtual and intact biological joints are shown to coordinate reinforces the theory that desired movement trajectories are mentally formulated in an abstract task space which does not depend on physical limb configurations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12984-021-00829-z.

Using an ankle robotic device for motor performance and motor learning evaluation

In this paper we performed the evaluation of ankle motor performance and motor learning during a goal-directed task, executed using the pediAnklebot robot. The protocol consisted of 3 phases (Familiarization, Adaptation, and Wash Out) repeated one time for each movement direction (plantarflexion, dorsiflexion, inversion, and eversion). During Familiarization and Wash out subjects performed goal-directed movements in unperturbed environment, whereas during Adaptation phase, a curl viscous force field was applied and it was randomly removed 10 times out of 200. Ankle motor performance was evaluated by means of a set of indices grouped into: accuracy, smoothness, temporal, and stopping indices. Learning Index was calculated to study the motor learning during the adaptation phase, which was subdivided into 5 temporal intervals (target sets). The outcomes related to the ankle motor performance highlighted that the best performance in terms of accuracy and smoothness of the trajectories was obtained in dorsiflexion movements in the sagittal plane, and in inversion rotations in the frontal plane. Differences between movement directions revealed an anisotropic behavior of the ankle joint. Results of the Learning index showed a capability of the subjects to rapidly adapt to a perturbed force field depending on the magnitude of the perceived field.

Quantifying Age-Related Differences of Ankle Mechanical Properties Using a Robotic Device

A deep analysis of ankle mechanical properties is a fundamental step in the design of an exoskeleton, especially if it is to be suitable for both adults and children. This study aims at assessing age-related differences of ankle properties using pediAnklebot. To achieve this aim, we enrolled 16 young adults and 10 children in an experimental protocol that consisted of the evaluation of ankle mechanical impedance and kinematic performance. Ankle impedance was measured by imposing stochastic torque perturbations in dorsi-plantarflexion and inversion-eversion directions. Kinematic performance was assessed by asking participants to perform a goaldirected task. Magnitude and anisotropy of impedance were computed using a multipleinput multiple-output system. Kinematic performance was quantified by computing indices of accuracy, smoothness, and timing. Adults showed greater magnitude of ankle impedance in both directions and for all frequencies, while the anisotropy was higher in children. By analyzing kinematics, children performed movements with lower accuracy and higher smoothness, while no differences were found for the duration of the movement. In addition, adults showed a greater ability to stop the movement when hitting the target. These findings can be useful to a proper development of robotic devices, as well as for implementation of specific training programs.

Comparison of kinematic and EMG parameters between unassisted, fixed- and adaptive-stiffness robotic-assisted ankle movements in post-stroke subjects

In this paper, we present an assist-as-needed scheme that effectively adapted the assistance provided by an ankle rehabilitation robot according to patient's participation and performance during therapeutic movements. We performed an error-based estimation of the ankle impedance as a valid measure of the patient participation. Then, we computed the amount of robotic assistance by three steps: normalization of the combined patient-robot stiffness, optimization of patientrobot interaction, and finally, adaptation of the level of the robotic assistance according to patient's performance while playing a serious game. Four post-stroke patients evaluated our methodology using an impedance controlled robotic system to assist alternated open-ended dorsi/plantarflexion movements in sitting position. Experimental results indicated that the proposed adaptive-stiffness method improves patient participation and performance compared to a fixed-stiffness assistive method and to an unassisted baseline. We also found that adaptive assistance could optimize the patient's muscular activity during movements. Our strategy effectively assisted with a lower stiffness allowing more kinematic variability in motions leaded by patient, decreasing the total amount of provided assistance without compromising the overall performance during therapy.

Virtual reality for pediatric neuro-rehabilitation: adaptive visual feedback of movement to engage the mirror neuron system

Sensorimotor therapy gives optimal results when patients are cognitively engaged into highly repetitive tasks, a goal that most children find hard to pursue. This paper presents the key developments of our ongoing effort to design an interactive rehabilitation environment that motivates physically impaired children throughout their therapy. The continuous motivation is achieved by the system adapting fundamental therapeutic components to the performance of each child. The relevant movement is mirrored to an animated character projected in front of the child. We speculate that the visual observation of one's own movements will activate the "mirror neuron system", a brain system underlying the human capacity to learn by imitation. Our rehabilitation algorithm personalizes the difficulty of the tasks by adapting the difficulty of reaching virtual targets on the animated environment through changing the visual gain between real and animated movements. To track the sensorimotor performance, we estimated the time required to reach a target. To give a proof of concept for the adaptation of the visual gain, we developed a serious game driven by a Leap Motion device. In addition to becoming a testbed for studying sensorimotor integration and neuroplasticity, the proposed notion of visual gain can be integrated into a highly engaging environment in which physically impaired children will play their way to recovery.

Testing the concurrent validity of a naturalistic upper extremity reaching task

Point-to-point reaching has been widely used to study upper extremity motor control. We have been developing a naturalistic reaching task that adds tool manipulation and object transport to this established paradigm. The purpose of this study was to determine the concurrent validity of a naturalistic reaching task in a sample of healthy adults. This task was compared to the criterion measure of standard point-to-point reaching. Twenty-eight adults performed unconstrained out-and-back movements in three different directions relative to constant start location along midline using their nondominant arm. In the naturalistic task, participants manipulated a tool to transport objects sequentially between physical targets anchored to the planar workspace. In the standard task, participants moved a digital cursor sequentially between virtual targets, veridical to the planar workspace. In both tasks, the primary measure of performance was trial time, which indicated the time to complete 15 reaches (five cycles of three reaches/target). Two other comparator tasks were also designed to test concurrent validity when components of the naturalistic task were added to the standard task. Spearman's rank correlation coefficients indicated minimal relationship between the naturalistic and standard tasks due to differences in progressive task difficulty. Accounting for this yielded a moderate linear relationship, indicating concurrent validity. The comparator tasks were also related to both the standard and naturalistic task. Thus, the principles of motor control and learning that have been established by the wealth of point-to-point reaching studies can still be applied to the naturalistic task to a certain extent.

Robot-Aided Neurorehabilitation: A Pediatric Robot for Ankle Rehabilitation

This paper presents the pediAnklebot, an impedance-controlled low-friction, backdriveable robotic device developed at the Massachusetts Institute of Technology that trains the ankle of neurologically impaired children of ages 6-10 years old. The design attempts to overcome the known limitations of the lower extremity robotics and the unknown difficulties of what constitutes an appropriate therapeutic interaction with children. The robot's pilot clinical evaluation is on-going and it incorporates our recent findings on the ankle sensorimotor control in neurologically intact subjects, namely the speed-accuracy tradeoff, the deviation from an ideally smooth ankle trajectory, and the reaction time. We used these concepts to develop the kinematic and kinetic performance metrics that guided the ankle therapy in a similar fashion that we have done for our upper extremity devices. Here we report on the use of the device in at least nine training sessions for three neurologically impaired children. Results demonstrated a statistically significant improvement in the performance metrics assessing explicit and implicit motor learning. Based on these initial results, we are confident that the device will become an effective tool that harnesses plasticity to guide habilitation during childhood.