Using Dynamic Bayesian Optimization to Induce Desired Effects in the Presence of Motor Learning: a Simulation Study.

Human-in-the-loop (HIL) optimization is a control paradigm used for tuning the control parameters of human-interacting devices while accounting for variability among individuals. A limitation of state-of-the-art HIL optimization algorithms such as Bayesian Optimization (BO) is that they assume that the relationship between control parameters and user response does not change over time. BO can be modified to account for the dynamics of the user response by implementing time into the kernel function, a method known as Dynamic Bayesian Optimization (DBO).

Robot-Aided Training of Propulsion: Effects of Torque Pulses Applied to the Hip and Knee Joint Under User-Driven Treadmill Control.

Objective: to establish whether torque pulses ap-plied by an exoskeleton to the hip and knee joint modulate propulsion mechanics and whether changes in propulsion me-chanics are sustained after exposure to torque pulses under user-driven treadmill control.

Methods: we applied twelve for-mulations of torque pulses consecutively over 300 strides to 22 healthy participants, and quantified the evolution of four outcome measures - gait speed (GS), hip extension (HE), trailing limb angle (TLA), normalized propulsive impulse (NPI) - before, during, and immediately after training.

Results: Metrics of propulsion mechanics significantly changed both during and after training.

Modeling Neuromotor Adaptation to Pulsed Torque Assistance During Walking.

Multiple mechanisms of motor learning contribute to the response of individuals to robot-aided gait training, including error-based learning and use-dependent learning. Previous models described either of these mechanisms, but not both, and their relevance to gait training is unknown. In this paper, we establish the validity of existing models to describe the response of healthy individuals to robot-aided training of propulsion via a robotic exoskeleton, and propose a new model that accounts for both use-dependent and error-based learning.

Development of an MRI-compatible robotic perturbation system for studying the task-dependent contribution of the brainstem to long-latency responses.

Methodological constraints have hindered direct measurement of reticulospinal tract (RST) function. The RST is thought to contribute to the increase in the amplitude of a long latency response (LLR), a stereotypical response evoked in stretched muscles, that arises when participants are asked to "resist" a perturbation. Thus, functional magnetic resonance imaging (fMRI) during robot-evoked LLRs under different task goals may be a method to measure motor-related RST function.

Individual Muscle Force Estimation in the Human Forearm Using Multi-Muscle MR Elastography (MM-MRE).

Objective: To establish the sensitivity of magnetic resonance elastography (MRE) to active muscle contraction in multiple muscles of the forearm.

Methods: We combined MRE of forearm muscles with an MRI-compatible device, the MREbot, to simultaneously measure the mechanical properties of tissues in the forearm and the torque applied by the wrist joint during isometric tasks. We measured shear wave speed of thirteen forearm muscles via MRE in a series of contractile states and wrist postures and fit these outputs to a force estimation algorithm based on a musculoskeletal model.

Concurrent Contribution of Co-contraction to Error Reduction during Dynamic Adaptation of the Wrist.

MRI-compatible robots provide a means of studying brain function involved in complex sensorimotor learning processes, such as adaptation. To properly interpret the neural correlates of behavior measured using MRI-compatible robots, it is critical to validate the measurements of motor performance obtained via such devices. Previously, we characterized adaptation of the wrist in response to a force field applied via an MRI-compatible robot, the MR-SoftWrist.

Changes in Resting State Functional Connectivity Associated with Dynamic Adaptation of Wrist Movements.

Dynamic adaptation is an error-driven process of adjusting planned motor actions to changes in task dynamics (Shadmehr, 2017). Adapted motor plans are consolidated into memories that contribute to better performance on re-exposure. Consolidation begins within 15 min following training (Criscimagna-Hemminger and Shadmehr, 2008), and can be measured via changes in resting state functional connectivity (rsFC).

Measurement of stretch-evoked brainstem function using fMRI.

Knowledge on the organization of motor function in the reticulospinal tract (RST) is limited by the lack of methods for measuring RST function in humans. Behavioral studies suggest the involvement of the RST in long latency responses (LLRs). LLRs, elicited by precisely controlled perturbations, can therefore act as a viable paradigm to measure motor-related RST activity using functional Magnetic Resonance Imaging (fMRI).

Effects of Perturbation Velocity, Direction, Background Muscle Activation, and Task Instruction on Long-Latency Responses Measured From Forearm Muscles.

The central nervous system uses feedback processes that occur at multiple time scales to control interactions with the environment. The long-latency response (LLR) is the fastest process that directly involves cortical areas, with a motoneuron response measurable 50 ms following an imposed limb displacement. Several behavioral factors concerning perturbation mechanics and the active role of muscles prior or during the perturbation can modulate the long-latency response amplitude (LLRa) in the upper limbs, but the interactions among many of these factors had not been systematically studied before.

Robot-Aided Training of Propulsion During Walking: Effects of Torque Pulses Applied to the Hip and Knee Joints During Stance.

We sought to evaluate the effects of the application of torque pulses to the hip and knee joint via a robotic exoskeleton in the context of training propulsion during walking. Based on our previous study, we formulated a set of conditions of torque pulses applied to the hip and knee joint associated with changes in push-off posture, a component of propulsion. In this work, we quantified the effects of hip/knee torque pulses on metrics of propulsion, specifically hip extension (HE) and normalized propulsive impulse (NPI), in two experiments.

Training Propulsion via Acceleration of the Trailing Limb.

Walking function, which is critical to performing many activities of daily living, is commonly assessed by walking speed. Walking speed is dependent on propulsion, which is governed by ankle moment and the posture of the trailing limb during push-off. Here, we present a new gait training paradigm that utilizes a dual belt treadmill to train both components of propulsion by accelerating the belt of the trailing limb during push-off.

StretchfMRI: a novel technique to quantify the contribution of the reticular formation to long-latency responses via fMRI.

Increased reticulospinal (RS) function has been observed to cause both positive and negative outcomes in the recovery of motor function after corticospinal lesions such as stroke. Current knowledge of RS function is limited by the lack of accurate, noninvasive methods for measuring RS function. Recent studies suggest that the RS tract may be involved in processing and generating Long Latency Responses (LLRs).

Identifying the neural representation of fast and slow states in force field adaptation via fMRI.

Although neurorehabilitation is centered on motor learning and control processes, our understanding of how the brain learns to control movement is still limited. Motor adaptation is an error-driven motor learning process that is amenable to study in the laboratory setting. Behavioral studies of motor adaptation have coupled clever task design with computational modeling to study the control processes that underlie motor adaptation.

Single-stride exposure to pulse torque assistance provided by a robotic exoskeleton at the hip and knee joints.

Robot assisted gait retraining is an increasingly common method for supporting restoration of walking function after neurological injury. Gait speed, an indicator of walking function, is correlated with propulsive force, a measure modulated by the posture of the trailing limb at push-off. With the ultimate goal of improving efficacy of robot assisted gait retraining, we sought to directly target gait propulsion, by exposing subjects to pulses of joint torque applied at the hip and knee joints to modulate push-off posture.

MM-MRE: a new technique to quantify individual muscle forces during isometric tasks of the wrist using MR elastography.

Non-invasive in-vivo measurement of individual muscle force is limited by the infeasibility of placing force sensing elements in series with the musculo-tendon structures. While different methods based either on shear wave elastography or electromyography have been recently proposed to non-invasively estimate individual muscle forces, they can only be used to quantity forces in a limited set of superficial muscles. As such, they are not suitable to study the neuromuscular control of movements that require coordinated action of multiple muscles.

Training propulsion: Locomotor adaptation to accelerations of the trailing limb.

Many stroke survivors suffer from hemiparesis, a condition that results in impaired walking ability. Walking ability is commonly assessed by walking speed, which is dependent on propulsive force generation both in healthy and stroke populations. Propulsive force generation is determined by two factors: ankle moment and the posture of the trailing limb during push-off.

Model-Based Estimation of Individual Muscle Force Based on Measurements of Muscle Activity in Forearm Muscles During Isometric Tasks.

Objective: Several forward dynamics estimation approaches have been proposed to estimate individual muscle force. However, characterization of the estimation error that arises when measurements are available only from a subset of the muscles involved in the movement under analysis, as is the case of the forearm muscles, has been limited. Our objectives were: first, to quantify the accuracy of forward-dynamics muscle force estimators for forearm muscles; and second, to develop a muscle force estimator that is accurate even when measurements are available only from a subset of muscles acting on a given joint or segment.

The effect of stride length on lower extremity joint kinetics at various gait speeds.

Robot-assisted training is a promising tool under development for improving walking function based on repetitive goal-oriented task practice. The challenges in developing the controllers for gait training devices that promote desired changes in gait is complicated by the limited understanding of the human response to robotic input. A possible method of controller formulation can be based on the principle of bio-inspiration, where a robot is controlled to apply the change in joint moment applied by human subjects when they achieve a gait feature of interest.

Model-Based Analysis of the Stiffness of the Wrist Joint in Active and Passive Conditions.

The control of joint stiffness is a fundamental mechanism used to control human movements. While many studies have observed how stiffness is modulated for tasks involving shoulder and elbow motion, a limited amount of knowledge is available for wrist movements, though the wrist plays a crucial role in manipulation. We have developed a computational framework based on a realistic musculoskeletal model, which allows one to calculate the passive and active components of the wrist joint stiffness.

Quantitative Testing of fMRI-Compatibility of an Electrically Active Mechatronic Device for Robot-Assisted Sensorimotor Protocols.

Objective: To develop a quantitative set of methods for testing the functional magnetic resonance imaging (fMRI) compatibility of an electrically-active mechatronic device developed to support sensorimotor protocols during fMRI.

Methods: The set of methods includes phantom and in vivo experiments to measure the effect of a progressively broader set of noise sources potentially introduced by the device. Phantom experiments measure the radio-frequency (RF) noise and temporal noise-to-signal ratio (tNSR) introduced by the device.

Toward goal-oriented robotic gait training: The effect of gait speed and stride length on lower extremity joint torques.

Robot-assisted gait training is becoming increasingly common to support recovery of walking function after neurological injury. How to formulate controllers capable of promoting desired features in gait, i.e.

Effects of Assist-As-Needed Upper Extremity Robotic Therapy after Incomplete Spinal Cord Injury: A Parallel-Group Controlled Trial.

Background: Robotic rehabilitation of the upper limb following neurological injury has been supported through several large clinical studies for individuals with chronic stroke. The application of robotic rehabilitation to the treatment of other neurological injuries is less developed, despite indications that strategies successful for restoration of motor capability following stroke may benefit individuals with incomplete spinal cord injury (SCI) as well. Although recent studies suggest that robot-aided rehabilitation might be beneficial after incomplete SCI, it is still unclear what type of robot-aided intervention contributes to motor recovery.

Kinesthetic Feedback During 2DOF Wrist Movements via a Novel MR-Compatible Robot.

We demonstrate the interaction control capabilities of the MR-SoftWrist, a novel MR-compatible robot capable of applying accurate kinesthetic feedback to wrist pointing movements executed during fMRI. The MR-SoftWrist, based on a novel design that combines parallel piezoelectric actuation with compliant force feedback, is capable of delivering 1.5 N [Formula: see text] of torque to the wrist of an interacting subject about the flexion/extension and radial/ulnar deviation axes.

Design of a series elastic actuator for a compliant parallel wrist rehabilitation robot.

This paper presents the design of a novel linear series elastic actuator purposely designed to match the requirements of robots for wrist rehabilitation: backdriveability, intrinsic compliance, and capability to be controlled as ideal force/torque sources. An existing rehabilitation robot is adapted to include intrinsic compliance in the design. A novel linear compliant element is designed to meet dimensional and force/stiffness requirements; a force sensing scheme involving a Hall-effect sensor is optimized in FEM simulations and developed.

Adaptive control of a serial-in-parallel robotic rehabilitation device.

Robotic rehabilitation is an effective platform for sensorimotor training after neurological injuries. In this paper, an adaptive controller is developed and implemented for the RiceWrist, a serial-in-parallel robot mechanism for upper extremity robotic rehabilitation. The model-based adaptive controller implementation requires a closed form dynamic model, valid for a restricted domain of generalized coordinates.