Defining a systems framework for characterizing physical work demands with wearable sensors.

Measuring the physical demands of work is important in understanding the relationship between exposure to these job demands and their impact on the safety, health, and well-being of working people. However, work is changing and our knowledge of job demands should also evolve in anticipation of these changes. New opportunities exist for noninvasive long-term measures of physical demands through wearable motion sensors, including inertial measurement units, heart rate monitors, and muscle activity monitors. Inertial measurement units combine accelerometers, gyroscopes, and magnetometers to provide continuous measurement of a segment's motion and the ability to estimate orientation in 3-dimensional space. There is a need for a system-thinking perspective on how and when to apply these wearable sensors within the context of research and practice surrounding the measurement of physical job demands. In this paper, a framework is presented for measuring the physical work demands that can guide designers, researchers, and users to integrate and implement these advanced sensor technologies in a way that is relevant to the decision-making needs for physical demand assessment. We (i) present a literature review of the way physical demands are currently being measured, (ii) present a framework that extends the International Classification of Functioning to guide how technology can measure the facets of work, (iii) provide a background on wearable motion sensing, and (iv) define 3 categories of decision-making that influence the questions that we can ask and measures that are needed. By forming questions within these categories at each level of the framework, this approach encourages thinking about the systems-level problems inherent in the workplace and how they manifest at different scales. Applying this framework provides a systems approach to guide study designs and methodological approaches to study how work is changing and how it impacts worker safety, health, and well-being.

An Augmented Reality Hand-Eye Sensorimotor Impairment Assessment for Spaceflight Operations.

INTRODUCTION: Following a transition from microgravity to a gravity-rich environment (e.g., Earth, Moon, or Mars), astronauts experience sensorimotor impairment, primarily from a reinterpretation of vestibular cues, which can impact their ability to perform mission-critical tasks. To enable future exploration-class missions, the development of lightweight, space-conscious assessments for astronauts transitioning between gravity environments without expert assistance is needed.METHODS: We examined differences in performance during a two-dimensional (2D) hand-eye multidirectional tapping task, implemented in augmented reality in subjects (N = 20) with and without the presence of a vestibular-dominated sensorimotor impairment paradigm: the binaural bipolar application of a pseudorandom galvanic vestibular stimulation (GVS) signal. Metrics associated with both the impairment paradigm and task performance were assessed.RESULTS: Medial-lateral sway during balance on an anterior-posterior sway-referenced platform with eyes closed was most affected by GVS (effect size: 1.2), in addition to anterior-posterior sway (effect size: 0.63) and the vestibular index (effect size: 0.65). During the augmented reality task, an increase in time to completion (effect size: 0.63), number of misses (effect size: 0.52), and head linear accelerations (effect size: 0.30) were found in the presence of the selected GVS waveform.DISCUSSION: Findings indicate that this multidirectional tapping task may detect emergent vestibular-dominated impairment (near landing day performance) in astronauts. Decrements in speed and accuracy indicate this impairment may hinder crews' ability to acquire known target locations while in a static standing posture. The ability to track these decrements can support mission operations decisions.Allred AR, Weiss H, Clark TK, Stirling L. An augmented reality hand-eye sensorimotor impairment assessment for spaceflight operations. Aerosp Med Hum Perform. 2024; 95(2):69-78.

Performance on a target acquisition task differs between augmented reality and touch screen displays.

Target acquisition tasks quantify human motor and perceptual abilities while performing discrete tasks to support interface design and sensorimotor assessments. This study investigated the effects of display, Touchscreen and Augmented Reality (AR), on a standardized 2D multidirectional target acquisition task. Thirty-two participants performed the target acquisition task with both modality types and at two indexes of difficulty. The touchscreen modality yielded improved performance over AR as measured by accuracy, precision, error rates, throughput, and movement time. Throughput using the nominal index of difficulty was 10.12 bits/s for touchscreen and 3.11 bits/s for AR. AR designers can use the results to improve performance when designing AR interfaces by selecting larger buttons when accuracy and efficiency are required and by embedding perception cues to button target surfaces such as depth and proximity cues.

Telerehabilitation use and experiences in occupational and physical therapy through the early stages of the COVID-19 pandemic.

Telehealth has helped to increase access to rehabilitative services such as occupational and physical therapy. The early COVID-19 pandemic amplified the need for remote access to care, and the rapid implementation of telehealth systems provided a unique opportunity to learn from clinicians' experiences adopting telehealth for telerehabilitation applications. To understand these experiences, a self-administered online survey was conducted to capture perspectives on ease of telerehabilitation use and adoption from occupational and physical therapists. The survey captured retrospective views on telerehabilitation use pre-pandemic as well as real-time perspectives on telerehabilitation during the early stages of the pandemic (July to August 2020). The survey gathered information on clinician demographics (N = 109), clinicians' experiences with adopting or utilizing telerehabilitation systems, and their perceptions on remotely performing cognitive, emotional, and physical assessments via video-conferencing (a common mode of telehealth). Responses demonstrated a modest increase in telerehabilitation as a care setting (rate increase from 3.4% to 19.3%), and telerehabilitation was more generally tried during the early stages of the pandemic (41 clinicians explicitly reported telerehabilitation use). However, technology access and acceptance remained low, with 38 clinicians (35%) expressing concerns that technology was ineffective or impractical, unavailable, not covered by insurance, or not desired by their patients. Video-conferencing technology was perceived as generally ill-equipped to support clinicians in performing remote assessment tasks. Physical assessment tasks were considered particularly difficult, with 55% of clinicians rating their ability to perform these tasks in the range of moderately difficult to unable to perform. To address these difficulties and better augment clinical care, clinicians require more robust assessment methods that may combine video, mobile, and wearable technologies that would be accessible to a patient at home. When designing future telerehabilitation tools, information captured through these modes must be task-relevant, standardized, and understandable to a remote clinician.

Ankle exoskeleton torque controllers based on soleus muscle models.

Powered exoskeletons are typically task-specific, but to facilitate their wider adoption they should support a variety of tasks, which requires generalizeable controller designs. In this paper, we present two potential controllers for ankle exoskeletons based on soleus fascicles and Achilles tendon models. The methods use an estimate of the adenosine triphosphate hydrolysis rate of the soleus based on fascicle velocity. Models were evaluated using muscle dynamics from the literature, which were measured with ultrasound. We compare the simulated behavior of these methods against each other and to human-in-the-loop optimized torque profiles. Both methods generated distinct profiles for walking and running with speed variations. One of the approaches was more appropriate for walking, while the other approach estimated profiles similar to the literature for both walking and running. Human-in-the-loop methods require long optimizations to set parameters per individual for each specific task, the proposed methods can produce similar profiles, work across walking and running, and be implemented with body-worn sensors without requiring torque profile parameterization and optimization for every task. Future evaluations should examine how human behavior changes due to external assistance when using these control models.

Walking speed and dual task input modality impact performance on a self-paced treadmill.

Interference between a walking task (target speeds on a self-paced treadmill) and dual visual and tactile-visual response time task was investigated. Ambulatory dual-task scenarios reveal how attention is divided between walking and additional tasks, but the impact of walking speed and dual-task modality on gait characteristics and dual-task performance is unclear. The purpose of this study was to evaluate the effect of visual and tactile-visual dual-task on gait performance. Participants (n=15) targeted four speeds (0.5, 1.0, 1.3, and 1.5 m/s) on a self-paced treadmill with a visual speed indicator (a green region centered at the target speed). Participants completed the same speed profile on the treadmill without (Self-Paced) and with a response time dual task (Self-Paced with Dual Task) requiring finger-tap responses to go/no-go cues. Six gait characteristics were calculated: proportion of time in the desired speed green region (GTP), speed ratio (ratio of mean to target speed), time to green region after target speed change (NRT), normalized stride width (NSW), normalized stride length (NSL), and stride time (ST). Both stride length and width were normalized by participant leg length. Lower GTP and greater speed ratio at slower speeds during dual tasking indicate speed-dependent changes in gait characteristics. Changes in NSL and ST were more affected by speed than dual task. These findings support that when speed is a parameter that is tracked, participants do not universally decrease speed in the presence of a dual task. These findings can support the decisions made when designing new wearable technologies that support navigation, communication, and mobility.

Comparison of Display Modality and Human-in-the-Loop Presence for On-Orbit Inspection of Spacecraft.

OBJECTIVE: To investigate the impact of interface display modalities and human-in-the-loop presence on the awareness, workload, performance, and user strategies of humans interacting with teleoperated robotic systems while conducting inspection tasks onboard spacecraft. BACKGROUND: Due to recent advancements in robotic technology, free-flying teleoperated robot inspectors are a viable alternative to extravehicular activity inspection operations. Teleoperation depends on the user's situation awareness; consequently, a key to successful operations is practical bi-directional communication between human and robot agents. METHOD: Participants (n = 19) performed telerobotic inspection of a virtual spacecraft during two degrees of temporal communication, a Synchronous Inspection task and an Asynchronous Inspection task. Participants executed the two tasks while using three distinct visual displays (2D, 3D, AR) and accompanying control systems. RESULTS: Anomaly detection performance was better during Synchronous Inspection than the Asynchronous Inspection of previously captured imagery. Users' detection accuracy reduced when given interactive exocentric 3D viewpoints to accompany the egocentric robot view. The results provide evidence that 3D projections, either demonstrated on a 2D interface or augmented reality hologram, do not affect the mean clearance violation time (local guidance performance), even though the subjects perceived a benefit. CONCLUSION: In the current implementation, the addition of augmented reality to a classical egocentric robot view for exterior inspection of spacecraft is unnecessary, as its margin of performance enhancement is limited in comparison. APPLICATION: Results are presented to inform future human-robot interfaces to support crew autonomy for deep space missions.

Does the fit of personal protective equipment affect functional performance? A systematic review across occupational domains.

OBJECTIVE: To explore the effect of personal protective equipment (PPE) fit on functional performance across a range of occupational domains. BACKGROUND: PPE introduces an ergonomic, human systems integration, and mass burden to the wearer, and these factors are thought to be amplified if PPE is ill-fitting. However, few studies have considered the role of fit (static, dynamic, and cognitive) when evaluating PPE-related performance detriments in occupational settings. METHOD: A systematic literature review was conducted to identify relevant studies, which were then critically appraised based on methodological quality and collated to compare key findings and present evidence-based recommendations for future research directions across a range of occupational domains. RESULTS: 16 published studies met the inclusion criteria, 88% of which found that the fit of PPE had a statistically significant effect on occupational performance. Poorly sized PPE resulted in slower or increased reaction time; decreased range of motion or mobility; decreased endurance or tolerance; decreased pulmonary function; and altered muscle activation. Limited research met the inclusion criteria and those that did had risks of bias in methodology quality. CONCLUSION: Future research evaluating the effect of PPE on performance in occupational settings should aim to recruit a more representative population; consider sex as a covariate; quantify and evaluate PPE fit and performance when integrated with all relevant equipment items; include outcome measures related to all three categories of fit (static, dynamic, cognitive); and assess performance of operationally relevant tasks.

Challenges in Quantifying Heel-Lift During Spacesuit Gait.

INTRODUCTION: Heel-lift is a subjectively reported fit issue in planetary spacesuit boot prototypes that has not yet been quantified. Inertial measurement units (IMUs) could quantify heel-lift but are susceptible to integration drift. This work evaluates the use of IMUs and drift-correction algorithms, such as zero-velocity (ZVUs) and zero-position updates (ZPUs), to quantify heel-lift during spacesuited gait.METHODS: Data was originally collected by Fineman et al. in 2018 to assess lower body relative coordination in the spacesuit. IMUs were mounted on the spacesuit lower legs (SLLs) and spacesuit operator's shank as three operators walked on a level walkway in three spacesuit padding conditions. Discrete wavelet transforms were used to identify foot-flat phase and heel-off for each step. Differences in heel-off timepoints were calculated in each step as a potential indicator of heel-lift, with spacesuit-delayed heel-off suggesting heel-lift. Average drift rates were estimated prior to and after applying ZVUs and ZPUs.RESULTS: Heel-off timepoint differences showed instances of spacesuit-delayed heel-off and instances of operator-delayed heel-off. Drift rates after applying ZVUs and ZPUs suggested an upper time bound of 0.03 s past heel-off to measure heel-lift magnitude with an accuracy of 1 cm.DISCUSSION: Results suggest that IMUs may not be appropriate for quantifying the presence and magnitude of heel lift. Operator-delayed heel-off suggests that the SLL may be expanding prior to heel-off, creating a false vertical acceleration signal interpreted by this study to be spacesuit heel-off. Quantifying heel-off will therefore require improvements in IMU mounting to mitigate the effects of SLL, or alternative sensor technologies.Boppana A, Priddy ST, Stirling L, Anderson AP. Challenges in quantifying heel-lift during spacesuit gait. Aerosp Med Hum Perform. 2022; 93(8):643-648.

Users Maintain Task Accuracy and Gait Characteristics During Missed Exoskeleton Actuations Through Adaptations In Joint Kinematics.

In operational settings, lower-limb active exoskeletons may experience errors, where an actuation that should be present is missed. These missed actuations may impact users' trust in the system and the adapted human-exoskeleton coordination strategies. In this study, we introduced pseudorandom catch trials, in which an assistive exoskeleton torque was not applied, to understand the immediate responses to missed actuations and how users' internal models to an exoskeleton adapt upon repeated exposure to missed actuations. Participants (N = 15) were instructed to complete a stepping task while wearing a bilateral powered ankle exoskeleton. Human-exoskeleton coordination and trust were inferred from task performance (step accuracy), step characteristics (step length and width), and joint kinematics at selected peak locations of the lower limb. Step characteristics and task accuracy were not impacted by the loss of exoskeleton torque as hip flexion was modulated to support completing the stepping task during catch trials, which supports an impacted human-exoskeleton coordination. Reductions in ankle plantarflexion during catch trials suggest user adaptation to the exoskeleton. Trust was not impacted by catch trials, as there were no significant differences in task performance or gait characteristics between earlier and later strides. Understanding the interactions between human-exoskeleton coordination, task accuracy, and step characteristics will support development of exoskeleton controllers for non-ideal operational settings.

Impact of Haptic Cues and an Active Ankle Exoskeleton on Gait Characteristics.

OBJECTIVE: This study examined the interaction of gait-synchronized vibrotactile cues with an active ankle exoskeleton that provides plantarflexion assistance. BACKGROUND: An exoskeleton that augments gait may support collaboration through feedback to the user about the state of the exoskeleton or characteristics of the task. METHODS: Participants (N = 16) were provided combinations of torque assistance and vibrotactile cues at pre-specified time points in late swing and early stance while walking on a self-paced treadmill. Participants were either given explicit instructions (N = 8) or were allowed to freely interpret (N=8) how to coordinate with cues. RESULTS: For the free interpretation group, the data support an 8% increase in stride length and 14% increase in speed with exoskeleton torque across cue timing, as well as a 5% increase in stride length and 7% increase in speed with only vibrotactile cues. When given explicit instructions, participants modulated speed according to cue timing-increasing speed by 17% at cues in late swing and decreasing speed 11% at cues in early stance compared to no cue when exoskeleton torque was off. When torque was on, participants with explicit instructions had reduced changes in speed. CONCLUSION: These findings support that the presence of torque mitigates how cues were used and highlights the importance of explicit instructions for haptic cuing. Interpreting cues while walking with an exoskeleton may increase cognitive load, influencing overall human-exoskeleton performance for novice users. APPLICATION: Interactions between haptic feedback and exoskeleton use during gait can inform future feedback designs to support coordination between users and exoskeletons.

Assessment of a powered ankle exoskeleton on human stability and balance.

Wearable robotic systems, such as exoskeletons, are designed to assist human motion; however, they are typically only studied during level walking. Before exoskeletons are broadly integrated into unstructured environments, it will be important to evaluate exoskeletons in a broader set of relevant tasks. A balance beam traverse was used to represent a constrained foot placement task for examining balance and stability. Participants (n = 17) completed the task in their own shoes (Pre-Exoskeleton and Post-Exoskeleton trials), and when wearing a lower-limb exoskeleton (Dephy ExoBoot) in both powered and unpowered states. Data were collected via inertial measurement units (on the torso and feet) and analyzed on a pooled level (with data from all participants) and on an individual level (participant-specific confidence intervals). When examining pooled data, it was observed that the exoskeleton had mixed effects on stride stability metrics. When compared to the Post-Exoskeleton shoe control, it was observed that stride duration was increased when wearing the exoskeleton (both powered and unpowered states), while normalized stride length and stride speed were not affected. Despite the changes in stride stability, overall balance (as measured by torso sway) remained unaffected by exoskeleton state. On an individual level, it was observed that not all participants followed these general trends, and within each metric, some increased, some decreased, and some had no change in the Powered Exoskeleton condition when compared to the Post-Exoskeleton Shoe condition: normalized stride length (0% increased, 12% decreased, 88% no change), stride duration (35% increased, 0% decreased, 65% no change), and torso sway (0% increased, 12% decreased, 88% no change). Our findings suggest that the lower-limb exoskeleton evaluated can be used during tasks that require balancing, and we recommend that balancing tasks be included in standards for exoskeleton evaluation.

Body-Worn IMU-Based Human Hip and Knee Kinematics Estimation during Treadmill Walking.

Traditionally, inertial measurement unit (IMU)-based human joint angle estimation techniques are evaluated for general human motion where human joints explore all of their degrees of freedom. Pure human walking, in contrast, limits the motion of human joints and may lead to unobservability conditions that confound magnetometer-free IMU-based methods. This work explores the unobservability conditions emergent during human walking and expands upon a previous IMU-based method for the human knee to also estimate human hip angles relative to an assumed vertical datum. The proposed method is evaluated (N=12) in a human subject study and compared against an optical motion capture system. Accuracy of human knee flexion/extension angle (7.87∘ absolute root mean square error (RMSE)), hip flexion/extension angle (3.70∘ relative RMSE), and hip abduction/adduction angle (4.56∘ relative RMSE) during walking are similar to current state-of-the-art self-calibrating IMU methods that use magnetometers. Larger errors of hip internal/external rotation angle (6.27∘ relative RMSE) are driven by IMU heading drift characteristic of magnetometer-free approaches and non-hinge kinematics of the hip during gait, amongst other error sources. One of these sources of error, soft tissue perturbations during gait, is explored further in the context of knee angle estimation and it was observed that the IMU method may overestimate the angle during stance and underestimate the angle during swing. The presented method and results provide a novel combination of observability considerations, heuristic correction methods, and validation techniques to magnetic-blind, kinematic-only IMU-based skeletal pose estimation during human tasks with degenerate kinematics (e.g., straight line walking).

Actuation Timing Perception of a Powered Ankle Exoskeleton and Its Associated Ankle Angle Changes During Walking.

Robotic ankle exoskeletons have the potential to extend human ability, and actuation timing serves as one of the critical parameters in its controller design. While many experiments have investigated the optimal actuation timing values to achieve different objective functions (e.g. minimizing metabolic cost), studies on users' perception of control parameters are gaining interest as it gives information on people's comfort, coordination, and trust in using devices, as well as providing foundations on how the sensorimotor system detects the exoskeleton behavior changes. The purpose of this study was to evaluate people's sensitivity to changes in exoskeleton actuation timing and its associated exoskeleton ankle angle changes during walking. Participants (n =15) with little or no prior experience with ankle exoskeletons were recruited and performed a psychophysical experiment to characterize their just-noticeable difference (JND) thresholds for actuation timing. Participants wore a bilateral active ankle exoskeleton and compared pairs of torque profiles with different actuation timings and low peak torque (0.225 Nm/kg) while walking on the treadmill. The mean timing JND across participants was 2.8±0.6% stride period. Individuals exhibited different sensitivity towards actuation timing, and their associated exoskeleton ankle angle changes also varied. The variance in ankle angle changes might be explained by their differences in ankle stiffness and different ankle torques provided during walking. The results provide insights into how people perceive the changes in exoskeleton control parameters and show individual differences in exoskeleton usage. The actuation timing JND found in this study can also help determine the necessary controller precision.

Alternative cue and response modalities maintain the Simon effect but impact task performance.

Inhibitory control, the ability to inhibit impulsive responses and irrelevant stimuli, enables high level functioning and activities of daily living. The Simon task probes inhibition using interfering stimuli, i.e., cues spatially presented on the opposite side of the indicated response; incongruent response times (RT) are slower than congruent RTs. Operational applicability of the Simon task beyond finger/hand manipulations and visual/auditory cues is unclear, but important to consider as new technologies provide tactile cues and require motor responses from the lower extremity (e.g., exoskeletons). In this study, twenty participants completed the Simon task under four conditions, each combination of two cue (visual/tactile) and response (upper/lower-extremity) modalities. RT were significantly longer for incongruent than congruent cues across cue-response pairs. However, alternative cue-response pairs yielded slower RT and decreased accuracy for tactile cues and lower-extremity responses. Results support operational usage of the Simon task to probe inhibition using tactile cues and lower-extremity responses relevant for new technologies like exoskeletons and immersive environments.

Analysis of Flexible Ureteroscopic Motion and Kinematic Efficiency: A Simulation-Based Pilot Study.

Introduction: Flexible ureteroscopy (fURS) is the most common surgical procedure for treatment of urolithiasis. Various surgical disciplines and subspecialties have examined surgeon kinematics to improve assessment and generate measures of skill. Despite frequency of utilization, there is no undisputed method for evaluating fURS skills. Our pilot study utilized kinematic evaluations of fURS simulation to determine whether specific surgeon movements, techniques, and strategies correlate with measures of ureteroscopic (URS) efficiency. Methods: A motion capture system and standard video camera were employed to characterize surgeon movement variables. A URS simulation box was used by practicing urologists at various skill levels to perform a series of simple and complex URS movement tasks. Two tasks were chosen for this initial pilot analysis. Body kinematics, time to task completion, and URS movements were analyzed. Task efficiency was defined as quicker time to task completion and smaller ureteroscope end effector travel distance. A combined performance efficiency score (PES) was calculated using the root sum square of these two measures. Results: Twelve practicing urologists were enlisted. Average urologist age was 37 years with an average of 10.1 years of training; 50% were women, 50% were residents; and 33% had completed an Endourology fellowship. For the simple task, no kinematic data correlated with PES; for the complex task, participant head and torso movement correlated with PES (r = 0.60, p = 0.04 for head; r = 0.65, p = 0.02 for torso), with decreased body movement associated with higher efficiency. Conclusion: Our findings suggest that movement economy measures are associated with efficient URS manipulation for complex tasks. Decreased head and torso movement were associated with higher efficiency, suggesting that excess body movement may signal extraneous or improper URS movements. Additional assessment of these variables, including analysis in a clinical setting, is warranted as this may serve as a basis for improvement in endoscopic training and evaluation.

Individuals differ in muscle activation patterns during early adaptation to a powered ankle exoskeleton.

Exoskeletons have the potential to assist users and augment physical ability. To achieve these goals across users, individual variation in muscle activation patterns when using an exoskeleton need to be evaluated. This study examined individual muscle activation patterns during walking with a powered ankle exoskeleton. 60% of the participants were observed to reduce medial gastrocnemius activation with exoskeleton powered and increase with the exoskeleton unpowered during stance. 80% of the participants showed a significant increase in tibialis anterior activation upon power addition, with inconsistent changes upon power removal during swing. 60% of the participants that were able to adapt to the system, did not de-adapt after 5 min. Muscle activity patterns differ between individuals in response to the exoskeleton power state, and affected the antagonist muscle behavior during this early adaptation. It is important to understand these different individual behaviors to inform the design of exoskeleton controllers and training protocols.

Perception of Powered Ankle Exoskeleton Actuation Timing During Walking: A Pilot Study.

Actuation timing is an important parameter in powered ankle exoskeleton control that can significantly influence user experience and human-system performance. Previous studies have investigated the actuation timing through optimization under different objective functions, such as minimizing metabolic cost. However, little is known about people's psychological sense of actuation timing. This pilot study measured two subjects' sensitivity to small changes in actuation timing during walking. The just-noticeable difference (JND) threshold was determined via a fitted psychometric function, which quantified subjects' performance in discriminating between a pair of actuation timings. Subjects could detect changes of 3.6% and 6.8% stride period in actuation timing respectively, showing the difference in perception between individuals. The results from this pilot study provide a preliminary understanding of human perception towards exoskeleton control parameters, which offers insight on individual differences in exoskeleton usage and informs exoskeleton precision requirements to minimize undesired human-system interaction.

Utility of Inter-subject Transfer Learning for Wearable-Sensor-Based Joint Torque Prediction Models.

Generalizability between individuals and groups is often a significant hurdle in model development for human subjects research. In the domain of wearable-sensor-controlled exoskeleton devices, the ability to generalize models across subjects or fine-tune more general models to individual subjects is key to enabling widespread adoption of these technologies. Transfer learning techniques applied to machine learning models afford the ability to apply and investigate the viability and utility such knowledge-transfer scenarios. This paper investigates the utility of single- and multi-subject based parameter transfer on LSTM models trained for "sensor-to-joint torque" prediction tasks, with regards to task performance and computational resources required for network training. We find that parameter transfer between both single- and multi-subject models provide useful knowledge transfer, with varying results across specific "source" and "target" subject pairings. This could be leveraged to lower model training time or computational cost in compute-constrained environments or, with further study to understand causal factors of the observed variance in performance across source and target pairings, to minimize data collection and model retraining requirements to select and personalize a generic model for personalized wearable-sensor-based joint torque prediction technologies.