Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2023 Feb 10;18(2):e0270078. doi: 10.1371/journal.pone.0270078

Comparison of the dynamics of exoskeletal-assisted and unassisted locomotion in an FDA-approved lower extremity device: Controlled experiments and development of a subject-specific virtual simulator

Vishnu D Chandran 1, Sanghyun Nam 1, David Hexner 2, William A Bauman 3,4, Saikat Pal 1,5,*
Editor: Imre Cikajlo6
PMCID: PMC9916583  PMID: 36763637

Abstract

Robotic exoskeletons have considerable, but largely untapped, potential to restore mobility in individuals with neurological disorders, and other conditions that result in partial or complete immobilization. The growing demand for these devices necessitates the development of technology to characterize the human-robot system during exoskeletal-assisted locomotion (EAL) and accelerate robot design refinements. The goal of this study was to combine controlled experiments with computational modeling to build a virtual simulator of EAL. The first objective was to acquire a minimum empirical dataset comprising human-robot kinematics, ground reaction forces, and electromyography during exoskeletal-assisted and unassisted locomotion from an able-bodied participant. The second objective was to quantify the dynamics of the human-robot system using a subject-specific virtual simulator reproducing EAL compared to the dynamics of normal gait. We trained an able-bodied participant to ambulate independently in a Food and Drug Administration-approved exoskeleton, the ReWalk P6.0 (ReWalk Robotics, Yoknaem, Israel). We analyzed the motion of the participant during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers, with simultaneous measurements of (i) three-dimensional marker trajectories, (ii) ground reaction forces, (iii) electromyography, and (iv) exoskeleton encoder data. We created a virtual simulator in OpenSim, comprising a whole-body musculoskeletal model and a full-scale exoskeleton model, to determine the joint kinematics and moments during exoskeletal-assisted and unassisted maneuvers. Mean peak knee flexion angles of the human subject during exoskeletal-assisted walking were 50.1° ± 0.6° (left) and 52.6° ± 0.7° (right), compared to 68.6° ± 0.3° (left) and 70.7° ± 1.1° (right) during unassisted walking. Mean peak knee extension moments during exoskeletal-assisted walking were 0.10 ± 0.10 Nm/kg (left) and 0.22 ± 0.11 Nm/kg (right), compared to 0.64 ± 0.07 Nm/kg (left) and 0.73 ± 0.10 Nm/kg (right) during unassisted walking. This work provides a foundation for parametric studies to characterize the effects of human and robot design variables, and predictive modeling to optimize human-robot interaction during EAL.

Introduction

Wearable robotic exoskeletons have considerable potential to transform the people’s lives by reducing the metabolic cost of walking [1–3], carrying heavy loads [4, 5], or augmenting human performance to conduct strenous tasks over extended periods of time [6, 7]. The need for such assistive devices is particularly profound in restoring mobility in individuals with neurological disorders, including those of spinal cord injury, stroke, traumatic brain injury, and multiple sclerosis. Wearable robotic exoskeletons for rehabilitation of individuals with neurological disorders are relatively new. The ReWalk (ReWalk Robotics, Yoknaem, Israel) was the first lower extremity device to receive Food and Drug Administration (FDA) approval in 2014, followed by the Ekso (Ekso Bionics, Richmond, CA) and Indego (Parker Hannifin, Cleveland, OH) in 2016, and the Keeogo (B-temia Inc., Quebec City, Canada) in 2020. Prior studies have demonstrated the ability of individuals with neurological disorders who have partial impairments to those who are completely non-ambulatory to walk independently in robotic exoskeletons [8–19]. Exoskeletal-assisted locomotion (EAL) has been shown to improve functional and motor recovery [8], mobility [9, 10, 15, 16], chronic pain [20], muscle spasticity [20–22], cardiovascular health [13, 23], bowel function [24], bladder function [21, 22], and quality of life [21]. These studies highlight the growing demand for wearable robotic exoskeletons to improve physical and psychological health, employment opportunities, and community integration in persons with neurological disorders.

The growing demand for wearable robotic exoskeletons for rehabilitation of individuals with neurological disorders has exposed two gaps in knowledge. First, the dynamics (comprising joint kinematics and joint moments) of the human-robot system during EAL in FDA-approved devices are not well-understood. The human-robot entity during EAL represents a complex dynamic system, and parsing out the contributions of the human from the robot, together with their interaction effects, requires a minimum empirical dataset that includes human-robot kinematics, ground reaction forces, and electromyography (EMG). Such a dataset from EAL in an FDA-approved device does not exist. Kim and colleagues quantified the dynamics of the human-robot system during EAL in a device designed to augment human performance in industrial settings [25]; however, this exoskeleton is not FDA-approved for rehabilitation of individuals with neurological disorders.

It is important to study the dynamics of the human-robot system during EAL in FDA-approved devices because individuals with neurological disorders only have access to FDA-approved devices. Clinics and medical centers only use FDA-approved exoskeletons for rehabilitation of patients with neurological disorders. FDA approval is required to prescribe an exoskeleton for in home use by individuals with neurological disorders. Even in research settings, many institutional review boards require FDA approval for an exoskeleton to be used by participants with neurological disorders. These restrictions are due to the safety and regulatory concerns associated with the use of non-FDA approved devices, and highlight the importance of studying the dynamics of the human-robot system during EAL in FDA-approved devices. Previous studies with FDA-approved exoskeletons for neurological disorders have reported joint kinematics [10, 26–30], foot and ground reaction forces [14, 29, 30], and EMG [10, 17, 31, 32]. To the best of our knowledge, no prior study has reported joint moments of the human-robot system during EAL in an FDA-approved device. Joint moments are important because our understanding of the dynamics of a system is incomplete without joint moments. In dynamics, a moment of a force (for example, from a muscle) is a measure of its tendency to rotate a limb about its joint axis. Quantifying joint moments is mathematically complex since it requires solving equations of motions with subject-specific anthropometry, joint kinematics, and external forces as inputs. Furthermore, a minimum empirical dataset comprising joint kinematics, ground reaction forces, and EMG from a single session of controlled experiments with and without an FDA-approved exoskeleton is not available in the literature.

The second gap in knowledge exposed by the growing demand for wearable robotic exoskeletons is that existing technology to accelerate design refinements and improve human experience in these devices is extremely limited. It is not ethically feasible, and would undoubtedly be prohibitively expensive, to conduct a large number of human experiments to characterize the effects of an endless number of robot design variables, control strategies, and human parameters on the human-robot system during EAL. Computational simulation is a viable alternative to perform such parametric studies. Prior studies have used computational simulations of the human-robot system to optimize exoskeleton design [33–37], test control strategies [38–41], minimize the metabolic cost of locomotion [42–44], optimize assistance for pathological gait [45, 46], and study human-robot interaction [47–52]. However, no prior study has simulated EAL in an FDA-approved exoskeleton.

The goal of this study was to combine controlled experiments with computational modeling to build a virtual simulator of EAL in an FDA-approved lower extremity exoskeleton, the ReWalk P6.0. The first objective of this study was to acquire a minimum empirical dataset comprising human-robot kinematics, ground reaction forces, and EMG during exoskeletal-assisted and unassisted locomotion. The second objective of this study was to quantify the dynamics of the human-robot system using a subject-specific virtual simulator reproducing EAL compared to the dynamics of normal gait.

Methods

Participant recruitment

An able-bodied male (age 41 years, male, height 1.76 m, mass 89.4kg) was recruited for study participation. The participant was informed on all aspects of the study and signed informed consent in accordance with the policies of our Institutional Review Board. The Institutional Review Board at New Jersey Institute of Technology approved this study. The approval number is 2008001531R001.

Exoskeletal-assisted locomotion training

The participant was trained to perform sit-to-stand, stand-to-sit, and walking maneuvers in the ReWalk P6.0. Prior to initiating training, the exoskeleton was adjusted to fit the anthropometry of the participant, including pelvic band size, thigh leg length, shank leg length, knee bracket position, foot plate size, and ankle dorsiflexion setting [53]. Walking in the ReWalk P6.0 follows a standard procedure. To initiate walking in the ReWalk P6.0, the user starts in the standing position with the hand crutches on each side. The user activates the “Walk” mode through a controller watch, which is confirmed by a single beep. To perform the device-assisted gait maneuver, the user unweights each side in an alternating manner—that is, the right leg is unloaded by leaning the torso to the left, which is followed by leaning the torso to the right, which unloads the left leg, and so forth. The action of the user swaying from side-to-side and, thereby, unloading the contralateral leg, activates the hip and knee motors to initiate the stepping motion. The next several steps are a repetition of the back-and-forth swaying of the torso to unload the trailing leg to permit the stepping action to continue.

The sit-to-stand and stand-to-sit maneuvers in the ReWalk P6.0 follow standard procedures. To initiate the sit-to-stand maneuver, a user places the hand crutches posterior to the hip while seated. Then, the trainer activates the “Stand” mode through a controller, which is confirmed by a single beep. The hip motors flex the torso (~9°) to position the user’s upper body forward while they load the crutches. The user has to hold this position for 3 seconds, after which time the exoskeleton beeps thrice and activates the hip and knee motors to complete the sit-to-stand maneuver. Next, the stand-to-sit maneuver in the ReWalk P6.0 follows a similar procedure. A user places the hand crutches posterior to the hip while standing. Then, the trainer activates the “Sit” mode, which is confirmed by a single beep. The user gets into position by leaning backward (~6°) and loading the crutches. The user has to hold this position for 3 seconds, after which time the exoskeleton beeps thrice and activates the hip and knee motors to complete the stand-to-sit maneuver. By the end of the fourth 1-hour session, the participant was able to perform the sit-to-stand, stand-to-sit, and walking maneuvers in the exoskeleton independent of requiring any assistance.

Motion capture experiments during exoskeletal-assisted and unassisted locomotion

We analyzed the participant during exoskeletal-assisted and unassisted locomotion from a single motion capture session, including simultaneous measurements of three-dimensional (3-D) marker trajectories, ground reaction forces, EMG, and exoskeleton encoder data. A 16-camera motion capture system (Vicon V8 and Nexus, Vicon Motion Systems, Oxford, UK) was used to track retro-reflective markers at 100 Hz. The markers were placed on the participant and the exoskeleton to capture the position and orientation of all the human body and exoskeleton segments (Fig 1A and 1B). The markers were placed on the participant according to the Conventional Gait Model 2.5 template in Vicon Nexus [54]. Markers were placed on the exoskeleton based on a custom template (Fig 1A and 1B). Next, we recorded ground reaction forces from each foot using four overground force plates (Bertec Corp., Columbus, OH) sampled at 2000 Hz. The force data were filtered using a low-pass, fourth-order Butterworth filter with a cutoff frequency of 15 Hz.

Fig 1.

Fig 1

Open in a new tab

(A, B) Placement of retro-reflective markers on the human (blue, cyan, and red), exoskeleton (yellow), and crutches (green) to track the position and orientation of all the segments of the human-robot system. The blue and cyan markers represent the Conventional Gait Model (CGM) 2.5 marker set, with the cyan markers removed after initial calibration. The red markers are offset markers used to locate occluded anatomical landmarks when the participant is in the exoskeleton. During post-processing, the occluded anatomical landmarks are reconstructed using these offset markers to complete the CGM 2.5 marker set. Next, we created a custom template to track the different segments of the exoskeleton and crutches. (C) The human-robot model in the virtual simulator comprises a universal joint (six degrees of freedom) at the pelvic band to anchor the exoskeleton to the participant, and a pin joint (one degree of freedom) each at the hips and knees of the exoskeleton. The location of the different joint axes are shown. (D) A generic musculoskeletal model [55] was scaled to the participant’s anthropometry.

We recorded muscle EMG measurements from both legs using a 16-channel surface system (TrignoTM, Delsys Inc., Natick, MA) sampled at 2000 Hz. We measured EMG from rectus femoris, vastus lateralis, vastus medialis, semitendinosus, biceps femoris, gastrocnemius medialis, soleus, and tibialis anterior muscles using established protocols [56, 57]. The participant’s mean resting EMG was determined from unassisted standing trials with the participant instructed to remain stationary in a neutral pose with their muscles relaxed. The mean resting EMG value was subtracted from their raw EMG values from the locomotion trials to offset the data to zero. We filtered the EMG data using a fourth-order high-pass filter (30 Hz) to remove motion artifact and then full-wave rectified and filtered with a fourth-order Butterworth low-pass filter (6 Hz) to obtain the linear envelope of muscle activation [58]. The filtered EMG data were normalized to muscle-specific activations obtained from maximum voluntary contraction trials. To quantify muscle-specific activations from maximum voluntary contraction trials, a participant performed 5 trials of isometric muscle contractions for each muscle group to elicit maximum activation of the specific muscles [58]. For example, to elicit maximum activation of the quadriceps muscles, a participant sat on a chair with the knee at approximately 80° of flexion and was instructed to extend their knee against the resistance of the tester. Verbal encouragement was given to the participant to try to improve the effort with each trial. The peak EMG value from all 5 trials was assigned as a muscle’s maximum activation. Similar isometric muscle contraction trials were performed for the hamstrings, ankle plantarflexion, and ankle dorsiflexion muscle groups. In addition, we recorded exoskeleton encoder data from all motion capture trials, including timestamps corresponding to the activation of the motors and motor encoder angles. These data were synchronized with the 3-D marker trajectory, ground reaction force, and EMG data.

Motion capture of exoskeletal-assisted and unassisted walking

We acquired 3-D motion data while the participant walked back and forth over a 5.0 x 2.5 m even instrumented walkway with and without the exoskeleton. The participant walked 10 times back and forth with the exoskeleton at the participant’s preferred speed (0.47 ± 0.03 m/s), and 10 times unassisted at self-selected speed (1.11 ± 0.07 m/s). From this dataset, only successful trials were included for further analysis. A walking trial was determined as successful if the foot placements were entirely on single force plates, and there were no missing marker, ground reaction force, and EMG data. Additional criteria for a successful trial during exoskeletal-assisting walking was that the hand crutches were placed away from the force plates so as not to interfere with the ground reaction forces from the feet, and there were no missing exoskeleton encoder data. Based on these criteria, we obtained six successful trials each of exoskeletal-assisted and unassisted walking.

Motion capture of exoskeletal-assisted and unassisted sit-to-stand and stand-to-sit maneuvers

We acquired 3-D motion data while the participant performed sit-to-stand and stand-to-sit maneuvers with and without the exoskeleton. The participant used an armless piano bench with seat height of 43 cm positioned on the overground walkway that enabled each foot to be placed on a force plate during the entire duration of a maneuver. The participant performed the sit-to-stand and stand-to-sit maneuvers 10 times each with and without the exoskeleton. From this dataset, only successful trials were included for further analysis. A trial was determined as successful if the foot placements were entirely on single force plates and there were no missing marker, ground reaction force, and EMG data. Additional criteria for a successful trial during exoskeletal-assisting sit-to-stand/stand-to-sit were that the hand crutches were placed away from the force plates so as not to interfere with the ground reaction forces from the feet and there were no missing exoskeleton encoder data. Based on these criteria, we obtained four and five successful trials of exoskeletal-assisted and unassisted sit-to-stand maneuvers, respectively, and five successful trials each of exoskeletal-assisted and unassisted stand-to-sit maneuvers.

Virtual simulator reproducing exoskeletal-assisted and unassisted locomotion

We developed a virtual simulator in OpenSim [59] to reproduce exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers (Fig 1C and 1D). We integrated a previously published human musculoskeletal model [55] with a full-scale model of the ReWalk P6.0 exoskeleton. The musculoskeletal model comprised 24 segments and 37 degrees of freedom (DoFs): seven in each leg, six at the pelvis, three at the torso, and seven in each arm [55]. The seven DoFs in each leg included three DoFs at the ball-and-socket hip joint, a one DoF coupled knee mechanism with translations of the tibia and patella prescribed by the knee flexion angle, and one DoF revolute joints at the ankle, subtalar, and metatarsal joints. A six DoF universal joint at the pelvis was used to describe the pose of the musculoskeletal model with respect to the global origin. The three DoFs in the torso and upper body included a spherical joint connecting the torso to the pelvis. The seven DoFs in each arm included three DoFs at the ball-and-socket shoulder joint, one DoF revolute joint at the elbow, one DoF revolute joint between the radius and the ulna, and two DoFs universal joint between the radius and the hand for wrist flexion-extension and radial-ulnar deviation. The full-scale model of the exoskeleton comprised seven segments and four DoFs (Fig 1C). The four DoFs included a single DoF pin joint each at the hips and knees of the exoskeleton. The ankle joints in the exoskeleton were welded to the lower leg segments. Next, human-robot interactions were modeled using an additional six DoF universal joint at the pelvic band to anchor the exoskeleton to the participant (Fig 1C). The musculoskeletal model included 80 Hill-type muscle-tendon units [60]. The Hill-type muscle-tendon units captured the force-length-velocity properties of the lower extremity muscles, with muscle geometry and architecture based on adult cadaver data [61].

We adapted a previously published computational framework in OpenSim [55, 59] to determine the dynamics of exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. We scaled the generic musculoskeletal model to match the mass and segment lengths of the participant. We determined joint kinematics by performing Inverse Kinematics (IK) analyses in OpenSim. IK solves for kinematics by minimizing error between the experimentally measured marker positions and the corresponding markers on the human-robot model. We performed Inverse Dynamics (ID) analyses to compute net human-robot joint torques. For ID analyses, we simplified the human-robot model by incorporating the mass and inertial properties of the exoskeleton with the musculoskeletal segments, similar to previous studies [62, 63]. Exoskeleton masses were added to the respective segments in the musculoskeletal model. The centers of mass of the segments in the musculoskeletal model were moved to the locations corresponding to the combined human-robot segments.

Data analysis and statistical methods

We compared hip, knee, and ankle flexion-extension angles from each leg during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. For exoskeletal-assisted maneuvers, we determined the hip and knee flexion-extension angles from the robot using IK and exoskeleton encoder data; the RMS errors between these two methods were minimal (the worst case was 3.3°), as such only joint angles from IK are presented in this study. The joint angles from multiple trials were averaged for each leg. We compared the vertical, anterior-posterior (AP), and medial-lateral (ML) ground reaction forces from each leg during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. The ground reaction forces from EAL were normalized to the combined weight of the participant and the exoskeleton [30]; forces from unassisted locomotion were normalized to the participant’s body weight. Ground reaction forces from multiple trials were averaged for each leg. Next, we compared normalized EMG from the eight lower extremity muscles from each leg during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. Normalized EMG from multiple trials were averaged for each leg. Finally, we compared hip, knee, and ankle moments from each leg during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. The joint moments from EAL were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. The joint moments from multiple trials were averaged for each leg.

Results

Exoskeletal-assisted and unassisted walking

The virtual simulator reproduced exoskeletal-assisted and unassisted walking within acceptable tolerances, with average (±1 SD) RMS errors between experiment and simulator markers being 1.30 (± 0.05) cm and 1.20 (± 0.02) cm for exoskeletal-assisted and unassisted trials, respectively. Acceptable tolerance is less than 2 cm average RMS error between experiment and simulator markers, per OpenSim’s best practices [64, 65]. Our segment coordinate frames were obtained with reference to a single global coordinate frame. The initial values of the joint kinematics from the different segment coordinate frames were based on this single global coordinate frame. This provided a direct comparison of joint kinematics with and without the exoskeleton. We observed differences in joint kinematics between exoskeletal-assisted and unassisted walking (Fig 2). The participant’s hips were more flexed during exoskeletal-assisted walking compared to unassisted walking (Fig 2A and 2B). Mean peak hip extension angles of the human during exoskeletal-assisted walking were 3.1° ± 1.9° (left) and 0.2° ± 0.9° (right), compared to 27.2° ± 1.2° (left) and 25.1° ± 1.0° (right) during unassisted walking (Fig 2A and 2B). After subtracting the offset angles between the segment coordinate axes, average absolute difference between human and robot hip flexion-extension angles were 1.5° ± 1.4° (left) and 1.9° ± 1.0° (right). Next, mean peak knee flexion angles of the participant during exoskeletal-assisted walking were 50.1° ± 0.6° (left) and 52.6° ± 0.7° (right), compared to 68.6° ± 0.3° (left) and 70.7° ± 1.1° (right) during unassisted walking (Fig 2C and 2D). After subtracting the offset angles between the segment coordinate axes, average absolute difference between participant and robot mean knee flexion-extension angles were 1.7° ± 1.1° (left) and 0.8° ± 0.5° (right). Mean peak ankle plantarflexion angles of the participant during exoskeletal-assisted walking were 6.9° ± 2.8° (left) and 6.7° ± 3.9° (right), compared to 13.9° ± 2.5° (left) and 19.5° ± 1.9° (right) during unassisted walking (Fig 2E and 2F). Peak plantarflexion during exoskeletal-assisted walking occurred in early stance during braking, in contrast to peak plantarflexion at the end of propulsion (toe-off) during unassisted walking. We observed a minimal range of plantarflexion (5.3°-10.5°) in the propulsion phase during exoskeletal-assisted walking, compared to a larger range (27.1°- 30.3°) during unassisted walking.

Fig 2.

Fig 2

Open in a new tab

Average (±1 SD) hip (A, B), knee (C, D), and ankle (E, F) joint angles from six exoskeletal-assited (human: blue for left leg, red for right leg; robot: black) and six unassisted (human: grey) walking trials. The offset in angles between the participant (blue or red) and the robot (black) was due to different definitions of coordinate axes for each rigid body. The dashed vertical lines represent toe-off.

We observed differences in normalized ground reaction forces between exoskeletal-assisted and unassisted walking, especially during the braking and propulsion phases (Fig 3). Mean peak vertical ground reaction forces during braking were 0.91 ± 0.08 BW (left) and 0.89 ± 0.06 BW (right) during exoskeletal-assisted walking, compared to 1.09± 0.02 BW (left) and 1.05 ± 0.05 BW (right) during unassisted walking (Fig 3A and 3B). Mean peak vertical ground reaction forces during propulsion were 0.98 ± 0.02 BW (left) and 1.02 ± 0.02 BW (right) during exoskeletal-assisted walking, compared to 1.04 ± 0.04 BW (left) and 1.02 ± 0.02 BW (right) during unassisted walking (Fig 3A and 3B). Mean peak posterior ground reaction forces during braking were 0.08 ± 0.04 BW (left) and 0.08 ± 0.02 BW (right) during exoskeletal-assisted walking, compared to 0.17 ± 0.01 BW (left) and 0.17 ± 0.04 BW (right) during unassisted walking (Fig 3C and 3D). Mean peak anterior ground reaction forces during propulsion were 0.06 ± 0.02 BW (left) and 0.08 ± 0.06 BW (right) during exoskeletal-assisted walking, compared to 0.18 ± 0.01 BW (left) and 0.19 ± 0.01 BW (right) during unassisted walking (Fig 3C and 3D). Mean peak lateral ground reaction forces during braking were 0.07 ± 0.02 BW (left) and 0.06 ± 0.01 BW (right) during exoskeletal-assisted walking, compared to 0.09 ± 0.01 BW (left) and 0.09 ± 0.01 BW (right) during unassisted walking (Fig 3E and 3F). Mean peak lateral ground reaction forces during propulsion 0.06 ± 0.01 BW (left) and 0.07 ± 0.01 BW (right) during exoskeletal-assisted walking, compared to 0.08 ± 0.01 BW (left) and 0.10 ± 0.01 BW (right) during unassisted walking (Fig 3E and 3F). The decreased ground reaction forces during the braking and propulsion phases of exoskeletal-assisted walking were supported by EMG data (Fig 4). The tibialis anterior muscle, used for braking, was less active during the braking and terminal swing (to prepare for braking) phases during exoskeletal-assisted compared to unassisted walking (Fig 4O and 4P). The gastrocnemius medialis and soleus muscles, used for propulsion, were less active during the propulsion phase during exoskeletal-assisted compared to unassisted walking (Fig 4K–4N).

Fig 3.

Fig 3

Open in a new tab

Average (±1 SD) vertical (A, B), anterior-posterior (C, D) and medial-lateral (E, F) ground reaction forces from six exoskeletal-assisted and six unassisted walking trials. The ground reaction forces from EAL were normalized to the combined weight of the participant and the exoskeleton; forces from unassisted locomotion were normalized to the participant’s body weight (BW). The dashed vertical lines represent toe-off.

Fig 4. Average (+1 SD) normalized electromyography (EMG) data from six exoskeletal-assisted and six unassisted walking trials.

Fig 4

Open in a new tab

The EMG data were normalized using muscle-specific maximum voluntary contraction values. The dashed vertical lines represent toe-off.

We observed differences in joint moments between exoskeletal-assisted and unassisted walking (Fig 5). Mean peak hip flexion moments of the participant during exoskeletal-assisted walking 0.23 ± 0.05 Nm/kg (left) and 0.35 ± 0.06 Nm/kg (right), compared to 0.65 ± 0.07 Nm/kg (left) and 0.64 ± 0.09 Nm/kg (right) during unassisted walking (Fig 5A and 5B). Mean peak knee extension moments of the participant during exoskeletal-assisted walking were 0.10 ± 0.10 Nm/kg (left) and 0.22 ± 0.11 Nm/kg (right), compared to 0.64 ± 0.07 Nm/kg (left) and 0.73 ± 0.10 Nm/kg (right) during unassisted walking (Fig 5C and 5D). Mean peak ankle plantarflexion moments during exoskeletal-assisted walking were 0.80 ± 0.12 Nm/kg (left) and 0.78 ± 0.19 Nm/kg (right), compared to 1.29 ± 0.04 Nm/kg (left) and 1.12 ± 0.03 Nm/kg (right) during unassisted walking (Fig 5E and 5F).

Fig 5.

Fig 5

Open in a new tab

Average (±1 SD) hip (A, B), knee (C, D), and ankle (E, F) joint moments from six exoskeletal-assited (human: blue for left leg, red for right leg) and six unassisted (human: grey) walking trials. The joint moments from EAL were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. The dashed vertical lines represent toe-off.

Exoskeletal-assisted and unassisted sit-to-stand

The virtual simulator reproduced exoskeletal-assisted and unassisted sit-to-stand within acceptable tolerances, with average (±1 SD) RMS errors between experiment and simulator markers being 1.50 (± 0.04) cm and 1.90 (± 0.05) cm for exoskeletal-assisted and unassisted trials, respectively. Exoskeletal-assisted sit-to-stand in the ReWalk P6.0 follows a standard procedure that is different from unassisted sit-to-stand, including different temporal events and phases. Comparison of exoskeletal-assisted and unassisted sit-to-stand is included as supplemental information, including joint kinematics (S1 Fig), ground reaction forces (S2 Fig), EMG (S3 Fig), and joint moments (S4 Fig).

Exoskeletal-assisted and unassisted stand-to-sit

The virtual simulator reproduced exoskeletal-assisted and unassisted stand-to-sit within acceptable tolerances, with average (±1 SD) RMS errors between experiment and simulator markers being 1.40 (± 0.06) cm and 1.90 (± 0.12) cm for exoskeletal-assisted and unassisted trials, respectively. Exoskeletal-assisted stand-to-sit in the ReWalk P6.0 follows a standard procedure that is different from unassisted stand-to-sit, including different temporal events and phases. Comparison of exoskeletal-assisted and unassisted stand-to-sit is included as supplemental information, including joint kinematics (S5 Fig), ground reaction forces (S6 Fig), EMG (S7 Fig), and joint moments (S8 Fig).

Discussion

The goal of this study was to combine controlled experiments with computational modeling to build a virtual simulator of EAL in the ReWalk P6.0, an FDA-approved lower extremity exoskeleton for rehabilitation of patients with neurological disorders. The first objective of this study was to acquire a minimum empirical dataset comprising human-robot kinematics, ground reaction forces, and EMG during exoskeletal-assisted and unassisted locomotion. Here, we present a minimum empirical dataset from a single session of controlled experiments with an able-bodied participant performing exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit maneuvers. Our results provide a direct comparison of joint kinematics (Fig 2, S1 and S5 Figs), ground reaction forces (Fig 3, S2 and S6 Figs), and EMG (Fig 4, S3 and S7 Figs) during exoskeletal-assisted and unassisted locomotion in an FDA-approved exoskeleton. The second objective of this study was to quantify the dynamics of the human-robot system using a subject-specific virtual simulator reproducing EAL, and to compare with the dynamics of unassisted locomotion. The subject-specific virtual simulator reproduced EAL within acceptable tolerances (average±SD RMS errors of 1.20±0.02 to 1.30±0.05 cm for walking, 1.50±0.04 to 1.90±0.05 cm for sit-to-stand, and 1.40±0.06 to 1.90±0.12 cm for stand-to-sit maneuvers). Our results provide a direct comparison of joint moments during exoskeletal-assisted and unassisted locomotion in an FDA-approved exoskeleton (Fig 5, S4 and S8 Figs).

We are not aware of any prior literature that has provided a minimum empirical dataset to quantify the dynamics of the human-robot system during EAL in any FDA-approved exoskeleton, including the ReWalk P6.0. To the best of our knowledge, no prior study has reported a direct comparison of joint kinematics, ground reaction forces, EMG, and joint moments from a single session of controlled experiments with and without an FDA-approved exoskeleton. Our study advances previous work done with the ReWalk device in the following ways. First, three prior studies have quantified joint kinematics during exoskeletal-assisted walking in the ReWalk device [26, 29, 30]; our study builds on this prior work to provide a direct comparison of joint kinematics from exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit activities in the ReWalk device (Fig 2, S1 and S5 Figs). Second, Hayes et al. reported ground reaction forces [29, 30] and Fineberg et al. reported foot reaction forces (obtained from in-shoe pressure insoles, F-Scan, TekScan, Boston, MA) during exoskeletal-assisted walking in the ReWalk device [14]; our study builds on this prior work to provide a direct comparison of ground reaction forces from exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit activities in the ReWalk device (Fig 3, S2 and S6 Figs). Third, no prior study has reported EMG or joint moments from any activity in the ReWalk device; our EMG (Fig 4, S3 and S7 Figs) and joint moments (Fig 5, S4 and S8 Figs) results add to the body of knowledge and provide a minimum empirical dataset needed to quantify the dynamics of the human-robot system during exoskeletal-assisted and unassisted walking, sit-to-stand, and stand-to-sit activities in the ReWalk device.

We have also compared our findings with data that is available in the literature from other FDA-approved exoskeletons. Our joint kinematics results from exoskeletal-assisted walking in the ReWalk P6.0 are comparable to those reported by Chang et al. using the Ekso device [28]. Our mean peak hip extension, knee flexion, and ankle plantarflexion angles were 0.2°-3.1°, 50.1°-52.6°, and 6.7°-6.9°, respectively (Fig 2); in comparison, mean peak hip extension, knee flexion, and ankle plantarflexion angles reported in Chang et al. are ~3.0°-4.0°, ~48.0°-50.0°, and ~6.0°-9.0°, respectively [28]. Ekelem et al. reported median peak hip and knee flexion angles of ~2.0°-16.0° and ~23.0°-25.0°, respectively, during exoskeletal-assisted walking in the Indego device [27]. Because they do not report peak hip extension angles during stance, a direct comparison with our hip extension angle data is not possible. Their peak knee flexion values (~23.0°-25.0°) are lower than our values (50.1°-52.6°). This is likely because the two participants in their study had high levels of spasticity; with functional electrical stimulation, Ekelem et al. were able to increase the participants’ peak knee flexion angles to ~35.0°-47.0° [27]. Next, Bellitto et al. compared EMG data from three incomplete spinal cord injured participants during exoskeletal-assisted (with the Ekso device) and unassisted walking [31]. The timings of our EMG activity are in agreement, but our normalized magnitudes are lower than those reported in Bellitto et al. This is likely because of two reasons: 1) our results are from an able-bodied participant, compared to incomplete spinal cord injured participants in Bellitto et al.; and 2) we normalized our EMG to maximum voluntary contraction data, while Bellitto et al. used a dynamic normalization technique [66], where EMG data from each muscle were normalized to the maximum value over all trials with and without the exoskeleton. There is no prior study with any FDA-approved exoskeletons to compare our results from sit-to-stand and stand-to-sit maneuvers.

This study provides a much-needed empirical dataset to develop virtual simulators reproducing EAL in FDA-approved devices. To the best of our knowledge, there is not a single virtual simulator of EAL in any FDA-approved exoskeleton. Prior studies have developed virtual simulators of EAL in custom-built exoskeletons [46–49, 51, 67], but these devices are not FDA-approved and are likely many years away from widespread application for the rehabilitation of patients with neurological disorders. Prior studies have also developed virtual simulators of EAL based on idealized or imaginary exoskeletons [39, 42–44, 50, 52]. For example, Bianco et al. developed a virtual simulator to examine how multi-joint assistance affects the metabolic cost of exoskeletal-assisted and unassisted walking [39]. This simulator, however, was not based on a physical exoskeleton and empirical data from unassisted walking was used as model inputs, with the assumption that joint kinematics, ground reaction forces, and EMG do not change between exoskeletal-assisted and unassisted walking. They stated in their limitations that incorporating the kinematic adaptations due to an exoskeleton would alter their metabolic cost predictions. It is likely that Bianco et al. and other studies that developed virtual simulators based on idealized or imaginary exoskeletal-assistance did so because they did not have empirical data from EAL.

A potential limitation of this study is that the ReWalk P6.0 was designed for patients with spinal cord injury, but the results from this study were obtained from a single able-bodied participant. We are currently actively recruiting able-bodied and spinal cord injured participants to validate the findings of our study. The goal of this manuscript is to report our novel framework combining controlled experiments with computational modeling to build subject-specific virtual simulators of EAL reproducing walking, sit-to-stand, and stand-to-sit maneuvers. Developing this framework represents substantial effort. We have described our methods in sufficient details for the research community to reproduce our work. In addition, we will make all empirical data from this study freely available through an online repository. There is benefit to publishing our dataset at the present time to enable the modeling community can start building on our work to address a broad range of research questions. A second limitation of this study is that our EAL experiments did not measure the external crutch reaction forces or the interaction forces due to the knee brackets and straps that were used to secure the participant in the exoskeleton. These forces were also not included in the virtual simulator and, as such, their effects on our findings remain undefined. A third potential limitation is that we incorporated the mass and inertial properties of the exoskeleton with the musculoskeletal segments, similar to that reported in previous studies, and thus simplified our virtual simulator for ID analyses [62, 63]. A more thorough approach would have been to segregate the human and robot components, include human-robot interaction forces as input and to solve the human and robot kinematic chains separately.

This study provides a framework for determining the dynamics of the human-robot system during EAL in an FDA-approved exoskeleton using controlled experiments and a subject-specific virtual simulator. Our framework is applicable to exoskeletons that are not FDA-approved. We created a first-of-a-kind empirical dataset comprising joint kinematics, ground reaction forces, EMG, and joint moments from exoskeletal-assisted and unassisted locomotion. The virtual simulator will provide a low-risk and cost-effective platform for parametric studies of human-robot interaction during EAL, a necessary step for researchers and clinicians to characterize the effects of human factors on EAL and establish standards for safe and sustained use of robotic exoskeletons. Of practical significance, this virtual simulator model will provide a platform for exoskeleton companies to conduct rapid design-phase evaluations to accelerate device refinements. We invite investigators to build on our work to develop accurate virtual simulators reproducing EAL to address a broad range of questions, permitting the design and manufacture of exoskeletal devices that more closely replicate a normal gait pattern.

Supporting information

S1 Fig. Average (±1 SD) joint angles from four exoskeletal-assited (A-F; human: blue for left leg, red for right leg; robot: black) and five unassisted (G-L; human: blue for left leg, red for right leg) sit-to-stand trials.

The offset in angles between the human (blue or red) and the robot (black) was due to different definitions of coordinate axes for each rigid body. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver.

(TIF)

Click here for additional data file. (480.5KB, tif)
S2 Fig. Average (±1 SD) ground reaction forces from four exoskeletal-assited (A-C) and five unassisted (D-F) sit-to-stand trials.

The ground reaction forces from exoskeletal-assisted locomotion were normalized to the combined weight of the participant and the exoskeleton; forces from unassisted locomotion were normalized to the participant’s body weight (BW). (A-C) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver. A-P = Anterior-Posterior, M-L = Medial-Lateral.

(TIF)

Click here for additional data file. (173.1KB, tif)
S3 Fig. Average (+1 SD) normalized electromyography (EMG) data from four exoskeletal-assisted (A-H) and five unassisted (I-P) sit-to-stand trials.

The EMG data were normalized using muscle-specific maximum voluntary contraction values. (A-H) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver. RF = rectus femoris, VL = vastus lateralis, VM = vastus medialis, ST = semitendinosus, BF = biceps femoris, GM = gastrocnemius medialis, SOL = soleus, TIB = tibialis anterior.

(TIF)

Click here for additional data file. (180.1KB, tif)
S4 Fig. Average (±1 SD) joint moments from four exoskeletal-assited (A-F) and five unassisted (G-L) sit-to-stand trials.

The joint moments from exoskeletal-assisted locomotion were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver.

(TIF)

Click here for additional data file. (185.4KB, tif)
S5 Fig. Average (±1 SD) joint angles from five exoskeletal-assited (A-F; human: blue for left leg, red for right leg; robot: black) and five unassisted (G-L; human: blue for left leg, red for right leg) stand-to-sit trials.

The offset in angles between the human (blue or red) and the robot (black) was due to different definitions of coordinate axes for each rigid body. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver.

(TIF)

Click here for additional data file. (489.8KB, tif)
S6 Fig. Average (±1 SD) ground reaction forces from five exoskeletal-assited (A-C) and five unassisted (D-F) stand-to-sit trials.

The ground reaction forces from exoskeletal-assisted locomotion were normalized to the combined weight of the participant and the exoskeleton; forces from unassisted locomotion were normalized to the participant’s body weight (BW). (A-C) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver. A-P = Anterior-Posterior, M-L = Medial-Lateral.

(TIF)

Click here for additional data file. (161.5KB, tif)
S7 Fig. Average (+1 SD) normalized electromyography (EMG) data from five exoskeletal-assisted (A-H) and five unassisted (I-P) stand-to-sit trials.

The EMG data were normalized using muscle-specific maximum voluntary contraction values. (A-H) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver. RF = rectus femoris, VL = vastus lateralis, VM = vastus medialis, ST = semitendinosus, BF = biceps femoris, GM = gastrocnemius medialis, SOL = soleus, TIB = tibialis anterior.

(TIF)

Click here for additional data file. (161.9KB, tif)
S8 Fig. Average (±1 SD) joint moments from five exoskeletal-assited (A-F) and five unassisted (G-L) stand-to-sit trials.

The joint moments from exoskeletal-assisted locomotion were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver.

(TIF)

Click here for additional data file. (165.1KB, tif)

Acknowledgments

We thank Assaf Tzioni from ReWalk Robotics for his support with access to the exoskeleton encoder data. We thank Lina Alsauskaite from ReWalk Robotics for training the research team on safe use of the exoskeleton. ReWalk Robotics did not play any role in study design, participant recruitment, methodology, and data analysis.

Data Availability

All relevant data are available on Figshare: https://doi.org/10.6084/m9.figshare.21967874.v1.

Funding Statement

WAB and SP received support from the Department of Veterans Affairs (https://www.va.gov/) grant VA RR&D # 1 I01 RX003561-01A2. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  • 1.Mooney LM, Rouse EJ, Herr HM. Autonomous exoskeleton reduces metabolic cost of human walking during load carriage. J Neuroeng Rehabil. 2014;11:80. Epub 2014/06/03. doi: 10.1186/1743-0003-11-80 ; PubMed Central PMCID: PMC4036406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Collins SH, Wiggin MB, Sawicki GS. Reducing the energy cost of human walking using an unpowered exoskeleton. Nature. 2015;522(7555):212–5. Epub 2015/04/02. doi: 10.1038/nature14288 ; PubMed Central PMCID: PMC4481882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Malcolm P, Galle S, Derave W, De Clercq D. Bi-articular Knee-Ankle-Foot Exoskeleton Produces Higher Metabolic Cost Reduction than Weight-Matched Mono-articular Exoskeleton. Front Neurosci. 2018;12:69. Epub 2018/03/20. doi: 10.3389/fnins.2018.00069 ; PubMed Central PMCID: PMC5841020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Fontana M, Vertechy R, Marcheschi S, Salsedo F, Bergamasco M. The body extender: A full-body exoskeleton for the transport and handling of heavy loads. IEEE Robotics and Automation Magazine. 2014;21(4):34–44. doi: 10.1109/MRA.2014.2360287 [DOI] [Google Scholar]
  • 5.Wang T, Zhu Y, Zheng T, Sui D, Zhao S, Zhao J. PALExo: A parallel actuated lower limb exoskeleton for high-load carrying. IEEE Access. 2020;8:67250–62. doi: 10.1109/ACCESS.2020.2986357 [DOI] [Google Scholar]
  • 6.Thamsuwan O, Milosavljevic S, Srinivasan D, Trask C. Potential exoskeleton uses for reducing low back muscular activity during farm tasks. Am J Ind Med. 2020;63(11):1017–28. Epub 2020/09/15. doi: 10.1002/ajim.23180 . [DOI] [PubMed] [Google Scholar]
  • 7.Van Engelhoven L, Poon N, Kazerooni H, Rempel D, Barr A, Harris-Adamson C. Experimental evaluation of a shoulder-support exoskeleton for overhead work: Influences of peak torque amplitude, task, and tool mass. IISE Transactions on Occupational Ergonomics and Human Factors. 2019;7(3–4):250–63. doi: 10.1080/24725838.2019.1637799 [DOI] [Google Scholar]
  • 8.Tsai CY, Delgado AD, Weinrauch WJ, Manente N, Levy I, Escalon MX, et al. Exoskeletal-assisted walking during acute inpatient rehabilitation leads to motor and functional improvement in persons with spinal cord injury: a pilot study. Arch Phys Med Rehabil. 2020;101(4):607–12. Epub 2020/01/01. doi: 10.1016/j.apmr.2019.11.010 . [DOI] [PubMed] [Google Scholar]
  • 9.Hong E, Gorman PH, Forrest GF, Asselin PK, Knezevic S, Scott W, et al. Mobility Skills With Exoskeletal-Assisted Walking in Persons With SCI: Results From a Three Center Randomized Clinical Trial. Frontiers in Robotics and AI. 2020;7(93). doi: 10.3389/frobt.2020.00093 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ramanujam A, Momeni K, Husain SR, Augustine J, Garbarini E, Barrance P, et al. Mechanisms for improving walking speed after longitudinal powered robotic exoskeleton training for individuals with spinal cord injury. Conf Proc IEEE Eng Med Biol Soc. 2018;2018:2805–8. Epub 2018/11/18. doi: 10.1109/EMBC.2018.8512821 . [DOI] [PubMed] [Google Scholar]
  • 11.Asselin PK, Avedissian M, Knezevic S, Kornfeld S, Spungen AM. Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton. J Vis Exp. 2016;(112). Epub 2016/06/25. doi: 10.3791/54071 ; PubMed Central PMCID: PMC4927801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Yang A, Asselin P, Knezevic S, Kornfeld S, Spungen AM. Assessment of In-Hospital Walking Velocity and Level of Assistance in a Powered Exoskeleton in Persons with Spinal Cord Injury. Top Spinal Cord Inj Rehabil. 2015;21(2):100–9. doi: 10.1310/sci2102-100 ; PubMed Central PMCID: PMC4568091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Asselin P, Knezevic S, Kornfeld S, Cirnigliaro C, Agranova-Breyter I, Bauman WA, et al. Heart rate and oxygen demand of powered exoskeleton-assisted walking in persons with paraplegia. J Rehabil Res Dev. 2015;52(2):147–58. doi: 10.1682/JRRD.2014.02.0060 . [DOI] [PubMed] [Google Scholar]
  • 14.Fineberg DB, Asselin P, Harel NY, Agranova-Breyter I, Kornfeld SD, Bauman WA, et al. Vertical ground reaction force-based analysis of powered exoskeleton-assisted walking in persons with motor-complete paraplegia. J Spinal Cord Med. 2013;36(4):313–21. doi: 10.1179/2045772313Y.0000000126 ; PubMed Central PMCID: PMC3758528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Androwis GJ, Sandroff BM, Niewrzol P, Fakhoury F, Wylie GR, Yue G, et al. A pilot randomized controlled trial of robotic exoskeleton-assisted exercise rehabilitation in multiple sclerosis. Mult Scler Relat Disord. 2021;51:102936. Epub 2021/04/21. doi: 10.1016/j.msard.2021.102936 . [DOI] [PubMed] [Google Scholar]
  • 16.Karunakaran KK, Gute S, Ames GR, Chervin K, Dandola CM, Nolan KJ. Effect of robotic exoskeleton gait training during acute stroke on functional ambulation. NeuroRehabilitation. 2021. Epub 2021/04/06. doi: 10.3233/NRE-210010 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ramanujam A, Cirnigliaro CM, Garbarini E, Asselin P, Pilkar R, Forrest GF. Neuromechanical adaptations during a robotic powered exoskeleton assisted walking session. J Spinal Cord Med. 2017:1–11. doi: 10.1080/10790268.2017.1314900 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kerdraon J, Previnaire JG, Tucker M, Coignard P, Allegre W, Knappen E, et al. Evaluation of safety and performance of the self balancing walking system Atalante in patients with complete motor spinal cord injury. Spinal Cord Ser Cases. 2021;7(1):71. Epub 2021/08/06. doi: 10.1038/s41394-021-00432-3 ; PubMed Central PMCID: PMC8338982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Tamburella F, Lorusso M, Tramontano M, Fadlun S, Masciullo M, Scivoletto G. Overground robotic training effects on walking and secondary health conditions in individuals with spinal cord injury: systematic review. J Neuroeng Rehabil. 2022;19(1):27. Epub 2022/03/17. doi: 10.1186/s12984-022-01003-9 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Stampacchia G, Rustici A, Bigazzi S, Gerini A, Tombini T, Mazzoleni S. Walking with a powered robotic exoskeleton: Subjective experience, spasticity and pain in spinal cord injured persons. NeuroRehabilitation. 2016;39(2):277–83. doi: 10.3233/NRE-161358 . [DOI] [PubMed] [Google Scholar]
  • 21.Baunsgaard CB, Nissen UV, Brust AK, Frotzler A, Ribeill C, Kalke YB, et al. Exoskeleton gait training after spinal cord injury: An exploratory study on secondary health conditions. J Rehabil Med. 2018;50(9):806–13. Epub 2018/09/06. doi: 10.2340/16501977-2372 . [DOI] [PubMed] [Google Scholar]
  • 22.Juszczak M, Gallo E, Bushnik T. Examining the Effects of a Powered Exoskeleton on Quality of Life and Secondary Impairments in People Living With Spinal Cord Injury. Top Spinal Cord Inj Rehabil. 2018;24(4):336–42. Epub 2018/11/22. doi: 10.1310/sci17-00055 ; PubMed Central PMCID: PMC6241230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Evans N, Hartigan C, Kandilakis C, Pharo E, Clesson I. Acute Cardiorespiratory and Metabolic Responses During Exoskeleton-Assisted Walking Overground Among Persons with Chronic Spinal Cord Injury. Top Spinal Cord Inj Rehabil. 2015;21(2):122–32. Epub 2015/09/14. doi: 10.1310/sci2102-122 ; PubMed Central PMCID: PMC4568093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Sale P, Russo EF, Scarton A, Calabro RS, Masiero S, Filoni S. Training for mobility with exoskeleton robot in spinal cord injury patients: a pilot study. Eur J Phys Rehabil Med. 2018;54(5):745–51. Epub 2018/03/09. doi: 10.23736/S1973-9087.18.04819-0 . [DOI] [PubMed] [Google Scholar]
  • 25.Kim S, Srinivasan D, Nussbaum MA, Leonessa A. Human gait during level walking with an occupational whole-body powered exoskeleton: not yet a walk in the park. IEEE Access. 2021;9:47901–11. [Google Scholar]
  • 26.Talaty M, Esquenazi A, Briceno JE, editors. Differentiating ability in users of the ReWalk TM powered exoskeleton: An analysis of walking kinematics. 2013 IEEE 13th international conference on rehabilitation robotics (ICORR); 2013: IEEE. [DOI] [PubMed] [Google Scholar]
  • 27.Ekelem A, Goldfarb M. Supplemental Stimulation Improves Swing Phase Kinematics During Exoskeleton Assisted Gait of SCI Subjects With Severe Muscle Spasticity. Front Neurosci. 2018;12:374. Epub 2018/06/19. doi: 10.3389/fnins.2018.00374 ; PubMed Central PMCID: PMC5992413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Chang SH, Zhu F, Patel N, Afzal T, Kern M, Francisco GE. Combining robotic exoskeleton and body weight unweighing technology to promote walking activity in tetraplegia following SCI: A case study. J Spinal Cord Med. 2020;43(1):126–9. Epub 2018/10/20. doi: 10.1080/10790268.2018.1527078 ; PubMed Central PMCID: PMC7006789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hayes SC, White M, White HSF, Vanicek N. A biomechanical comparison of powered robotic exoskeleton gait with normal and slow walking: An investigation with able-bodied individuals. Clin Biomech (Bristol, Avon). 2020;80:105133. Epub 2020/08/11. doi: 10.1016/j.clinbiomech.2020.105133 . [DOI] [PubMed] [Google Scholar]
  • 30.Hayes SC, White M, Wilcox CRJ, White HSF, Vanicek N. Biomechanical differences between able-bodied and spinal cord injured individuals walking in an overground robotic exoskeleton. PLoS One. 2022;17(1):e0262915. Epub 2022/01/28. doi: 10.1371/journal.pone.0262915 ; PubMed Central PMCID: PMC8794144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bellitto A, Mandraccia S, Leoncini C, Rossi L, Gamba S, Massone A, et al., editors. Walking after incomplete spinal cord injury: changes in muscle activations due to training with a robotic powered exoskeleton. 2020 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob); 2020; New York City, NY: IEEE. [Google Scholar]
  • 32.Escalona MJ, Bourbonnais D, Goyette M, Duclos C, Gagnon DH. Wearable exoskeleton control modes selected during overground walking affect muscle synergies in adults with a chronic incomplete spinal cord injury. Spinal Cord Ser Cases. 2020;6(1):26. Epub 2020/04/26. doi: 10.1038/s41394-020-0269-6 ; PubMed Central PMCID: PMC7182571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Safavi S, Ghafari AS, Meghdari A, editors. Design of an optimum torque actuator for augmenting lower extremity exoskeletons in biomechanical framework. IEEE International Conference on Robotics and Biomimetics; 2011; Phuket, Thailand. [Google Scholar]
  • 34.Guan X, Ji L, Wang R, Huang W, editors. Optimization of an unpowered energy-stored exoskeleton for patients with spinal cord injury. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society; 2016; Orlando, FL. [DOI] [PubMed] [Google Scholar]
  • 35.Hegarty AK, Petrella AJ, Kurz MJ, Silverman AK. Evaluating the Effects of Ankle-Foot Orthosis Mechanical Property Assumptions on Gait Simulation Muscle Force Results. J Biomech Eng. 2017;139(3). Epub 2016/12/18. doi: 10.1115/1.4035472 . [DOI] [PubMed] [Google Scholar]
  • 36.Arch ES, Stanhope SJ, Higginson JS. Passive-dynamic ankle-foot orthosis replicates soleus but not gastrocnemius muscle function during stance in gait: Insights for orthosis prescription. Prosthet Orthot Int. 2016;40(5):606–16. Epub 2015/07/26. doi: 10.1177/0309364615592693 . [DOI] [PubMed] [Google Scholar]
  • 37.Manns P, Sreenivasa M, Millard M, Mombaur K. Motion optimization and parameter identification for a human and lower back exoskeleton model. IEEE Robotics and Automation Letters. 2017;2(3):1564–70. [Google Scholar]
  • 38.Cseke B, Uchida T, Doumit M. Simulating Ideal Assistive Strategies to Reduce the Metabolic Cost of Walking in the Elderly. IEEE Trans Biomed Eng. 2022;PP. Epub 2022/02/25. doi: 10.1109/TBME.2022.3153951 . [DOI] [PubMed] [Google Scholar]
  • 39.Bianco NA, Franks PW, Hicks JL, Delp SL. Coupled exoskeleton assistance simplifies control and maintains metabolic benefits: A simulation study. PLoS One. 2022;17(1):e0261318. Epub 2022/01/06. doi: 10.1371/journal.pone.0261318 ; PubMed Central PMCID: PMC8730392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Franks PW, Bianco NA, Bryan GM, Hicks JL, Delp SL, Collins SH, editors. Testing simulated assistance strategies on a hip-knee-ankle exoskeleton: a case study. 8th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob); 2020; New York: IEEE. [Google Scholar]
  • 41.Durandau G, Farina D, Asin-Prieto G, Dimbwadyo-Terrer I, Lerma-Lara S, Pons JL, et al. Voluntary control of wearable robotic exoskeletons by patients with paresis via neuromechanical modeling. J Neuroeng Rehabil. 2019;16(1):91. Epub 2019/07/19. doi: 10.1186/s12984-019-0559-z ; PubMed Central PMCID: PMC6637518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Dembia CL, Silder A, Uchida TK, Hicks JL, Delp SL. Simulating ideal assistive devices to reduce the metabolic cost of walking with heavy loads. PLoS One. 2017;12(7):e0180320. Epub 2017/07/13. doi: 10.1371/journal.pone.0180320 ; PubMed Central PMCID: PMC5507502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Uchida TK, Seth A, Pouya S, Dembia CL, Hicks JL, Delp SL. Simulating Ideal Assistive Devices to Reduce the Metabolic Cost of Running. PLoS One. 2016;11(9):e0163417. Epub 2016/09/23. doi: 10.1371/journal.pone.0163417 ; PubMed Central PMCID: PMC5033584. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Nguyen VQ, Umberger BR, Sup FC. Predictive Simulation of Human Walking Augmented by a Powered Ankle Exoskeleton. IEEE Int Conf Rehabil Robot; Jun 2019. p. 53–8. doi: 10.1109/ICORR.2019.8779368 [DOI] [PubMed] [Google Scholar]
  • 45.Lim B, Hyoung S, Lee J, Seo K, Jang J, Shim Y, editors. Simulating gait assistance of a hip exoskeleton: case studies for ankle pathologies. 2017 IEEE International Conference on Robotics and Automation; 2017. [Google Scholar]
  • 46.Lerner ZF, Damiano DL, Bulea TC. Computational modeling of neuromuscular response to swing-phase robotic knee extension assistance in cerebral palsy. J Biomech. 2019;87:142–9. Epub 2019/03/14. doi: 10.1016/j.jbiomech.2019.02.025 ; PubMed Central PMCID: PMC6534383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Cho K, Kim Y, Yi D, Jung M, Lee K. Analysis and evaluation of a combined human–exoskeleton model under two different constraints condition. International Summit on Human Simulation 2012; May 20–22; St. Pete Beach, FL2012. [Google Scholar]
  • 48.Ferrati F, Bortoletto R, Pagello E. Virtual Modelling of a Real Exoskeleton Constrained to a Human Musculoskeletal Model. Biomimetic and Biohybrid Systems. Berlin, Heidelberg: Springer; 2013. p. 96–107. [Google Scholar]
  • 49.Fournier BN, Lemaire ED, Smith AJJ, Doumit M. Modeling and Simulation of a Lower Extremity Powered Exoskeleton. IEEE transactions on neural systems and rehabilitation engineering: a publication of the IEEE Engineering in Medicine and Biology Society. 2018;26(8):1596–603. Epub 2018/07/14. doi: 10.1109/TNSRE.2018.2854605 . [DOI] [PubMed] [Google Scholar]
  • 50.Zhu Y, Zhang G, Zhang C, Liu G, Zhao J. Biomechanical modeling and load-carrying simulation of lower limb exoskeleton. Biomed Mater Eng. 2015;26 Suppl 1:S729–38. Epub 2015/09/26. doi: 10.3233/BME-151364 . [DOI] [PubMed] [Google Scholar]
  • 51.Torricelli D, Cortés C, Lete N, Bertelsen Á, Gonzalez-Vargas JE, del-Ama AJ, et al. A Subject-Specific Kinematic Model to Predict Human Motion in Exoskeleton-Assisted Gait. Frontiers in Neurorobotics. 2018;12(18). doi: 10.3389/fnbot.2018.00018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Khamar M, Edrisi M, Zahiri M. Human-exoskeleton control simulation, kinetic and kinematic modeling and parameters extraction. MethodsX. 2019;6:1838–46. Epub 2019/09/12. doi: 10.1016/j.mex.2019.08.014 ; PubMed Central PMCID: PMC6726754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.ReWalk. ReWalk 6.0 Personal System User Guide. ReWalk Robotics. 2020:1–97. [Google Scholar]
  • 54.Leboeuf F, Baker R, Barre A, Reay J, Jones R, Sangeux M. The conventional gait model, an open-source implementation that reproduces the past but prepares for the future. Gait Posture. 2019;69:235–41. Epub 2019/04/28. doi: 10.1016/j.gaitpost.2019.04.015 . [DOI] [PubMed] [Google Scholar]
  • 55.Rajagopal A, Dembia CL, DeMers MS, Delp DD, Hicks JL, Delp SL. Full-Body Musculoskeletal Model for Muscle-Driven Simulation of Human Gait. IEEE Trans Biomed Eng. 2016;63(10):2068–79. Epub 2016/07/09. doi: 10.1109/TBME.2016.2586891 ; PubMed Central PMCID: PMC5507211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Rutherford DJ, Hubley-Kozey CL, Stanish WD. Maximal voluntary isometric contraction exercises: a methodological investigation in moderate knee osteoarthritis. J Electromyogr Kinesiol. 2011;21(1):154–60. Epub 2010/10/12. doi: 10.1016/j.jelekin.2010.09.004 . [DOI] [PubMed] [Google Scholar]
  • 57.Perotto A, Delagi EF, Iazzetti J, Morrison D. Anatomical Guide for the Electromyographer. 4th ed. Springfield, IL: Charles C. Thomas; 2005. 345 p. [Google Scholar]
  • 58.Pal S, Draper CE, Fredericson M, Gold GE, Delp SL, Beaupre GS, et al. Patellar maltracking correlates with vastus medialis activation delay in patellofemoral pain patients. Am J Sports Med. 2011;39(3):590–8. Epub 2010/11/16. doi: 10.1177/0363546510384233 . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Delp SL, Anderson FC, Arnold AS, Loan P, Habib A, John CT, et al. OpenSim: open-source software to create and analyze dynamic simulations of movement. IEEE transactions on biomedical engineering. 2007;54(11):1940–50. Epub 2007/11/21. doi: 10.1109/TBME.2007.901024 . [DOI] [PubMed] [Google Scholar]
  • 60.Thelen DG, Anderson FC, Delp SL. Generating dynamic simulations of movement using computed muscle control. J Biomech. 2003;36(3):321–8. doi: 10.1016/s0021-9290(02)00432-3 . [DOI] [PubMed] [Google Scholar]
  • 61.Delp SL, Loan JP, Hoy MG, Zajac FE, Topp EL, Rosen JM. An interactive graphics-based model of the lower extremity to study orthopaedic surgical procedures. IEEE Trans Biomed Eng. 1990;37(8):757–67. doi: 10.1109/10.102791 . [DOI] [PubMed] [Google Scholar]
  • 62.Grabke EP, Masani K, Andrysek J. Lower Limb Assistive Device Design Optimization Using Musculoskeletal Modeling: A Review. ASME J Med Devices. 2019;13(4). [Google Scholar]
  • 63.Sartori M, Durandau G, Farina D, editors. Neuromusculoskeletal models of human-machine interaction in individuals wearing lower limb assistive technologies. Third International Conference on NeuroRehabilitation; 2017; Segovia, Spain. [Google Scholar]
  • 64.Falisse A, Van Rossom S, Gijsbers J, Steenbrink F, van Basten BJH, Jonkers I, et al. OpenSim Versus Human Body Model: A Comparison Study for the Lower Limbs During Gait. J Appl Biomech. 2018:1–7. Epub 2018/05/29. doi: 10.1123/jab.2017-0156 . [DOI] [PubMed] [Google Scholar]
  • 65.Hicks JL, Seth A, Hamner SR, Delp S. Simulation with OpenSim—best practices 2016. [12/01/2022]. Available from: https://simtk-confluence.stanford.edu:8443/display/OpenSim/Simulation+with+OpenSim+-+Best+Practices. [Google Scholar]
  • 66.Dixon PC, Jansen K, Jonkers I, Stebbins J, Theologis T, Zavatsky AB. Muscle contributions to centre of mass acceleration during turning gait in typically developing children: A simulation study. J Biomech. 2015;48(16):4238–45. Epub 2015/11/12. doi: 10.1016/j.jbiomech.2015.10.028 . [DOI] [PubMed] [Google Scholar]
  • 67.Jackson RW, Dembia CL, Delp SL, Collins SH. Muscle-tendon mechanics explain unexpected effects of exoskeleton assistance on metabolic rate during walking. J Exp Biol. 2017;220(Pt 11):2082–95. Epub 2017/03/28. doi: 10.1242/jeb.150011 . [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Imre Cikajlo

29 Nov 2022

PONE-D-22-15991Comparison of the dynamics of exoskeletal-assisted and unassisted locomotion in an FDA-approved lower extremity device: Controlled experiments and development of a subject-specific virtual simulatorPLOS ONE

Dear Dr. Pal,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jan 13 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Imre Cikajlo, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for providing the following information in your Ethics Statement: "The Institutional Review Board at New Jersey Institute of Technology approved this study. The approval number is 2008001531R001." We ask that you also amend your Methods section to include this statement.

3. Thank you for stating in your Funding Statement:

“WAB and SP received support from the Department of Veterans Affairs (https://www.va.gov/) grant VA RR&D # 1 I01 RX003561-01A2. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a well-written paper that describes an experiment and simulator with one subject performing walking and sit-to-stand motions with and without an exoskeleton. I believe the manuscript is eventually worthy of publication, though I have a number of major and minor comments.

MAJOR REVISION

1. The authors repeatedly emphasize that existing studies have not been performed on an FDA-approved exoskeleton, and that the novelty of the work is usage with an FDA-approved exoskeleton. However, it is unclear why the FDA approval represents a major difference. It seems like existing studies have achieved similar results with less popular exoskeletons, and most of the approaches used in existing studies can be approved to other exoskeletons with relatively little modification. The authors should better explain why the study is more than simply applying existing methods to a more popular exoskeleton.

2. Related to the above: the Discussion acknowledges that several studies have already been done with the ReWalk exoskeleton (which the authors are using). The authors should better characterize how their work advances previous ReWalk work rather than simply saying their results are in line with previous ReWalk results.

3. Why did the authors recruit only a single subject for a single session? It seems like there would be significant intersubject variability that would both make it difficult to compare exoskeleton vs. no-exoskeleton conditions as well as make it difficult to evaluate performance of a simulation model. Additionally, since this was an able-bodied subject who only participated for a few sessions, it seems like the authors could have fairly easily obtained a larger sample size.

4. The simulation model does not appear to be extensively validated - the authors show that it can fairly accurately reproduce actual trajectories, but this is done by training and testing on the same subject in the same session. Is this sufficient for a real simulator? I would have expected it to be validated by examining predictive ability in different situations. And when the authors say that the simulator reproduced actual trajectories "within acceptable tolerances", what is an acceptable tolerance?

MINOR REVISION

1. Is it really necessary to list inclusion/exclusion criteria when only a single participant was recruited? Authors list things like pregnancy and lactation being exclusion criteria, which seems irrelevant for a male participant. They also list minimum and maximum weight, which again seems irrelevant since a single participant was recruited and his weight is known.

2. Could the authors briefly describe how maximum voluntary contraction protocols were performed?

3. Why were there 4 trials of exoskeleton-assisted sit-to-stand and 5 trials of exoskeleton-unassisted sit-to-stand? Why not 5 for both?

Reviewer #2: The manuscript presents a dataset collected on a healthy subject performing exoskeleton-assisted vs unassisted locomotion using an FDA-approved exoskeleton (ReWalk P6.0), and a subject-specific virtual simulator implemented using OpenSim environment, to compute joint kinematics (from motion capture and simulation) and moments (from simulation) for hip, knee and ankle in assisted vs unassisted conditions. in addition, EMG data from 8 lower-limb muscles were recorded, as well as ground reaction forces using force plates.

The main novelty claimed by the authors is the availability of such a dataset and simulation that includes an FDA-approved exoskeleton comparing assisted and unassisted locomotion. Nevertheless, whereas the need for having such a dataset and simulation is well explained in the manuscript, it is not clear which is the added value of including an FDA-approved exoskeleton, with respect to other examples (cited by the authors in the manuscript) using different exoskeletons. In my opinion, this alone cannot be the main claim of the manuscript given that: no novel methods are presented; different exoskeletons might result in different human-robot interaction when assistance is delivered; other similar data in literature could be used for the simulator if the benefit would be the possibility to "conduct rapid design-phase evaluations to accelerate device refinements" (Discussion).

The authors have stated clearly the limitations of the study, including having no participants with altered gait patterns (e.g., spinal cord) and not separating human and robotic dynamics in the virtual simulator (one aspect that was described in the Introduction as particularly important and not very well explored in the state of the art). Exploring further more at least one of these limitations would be important to justify the significance of the manuscript. Indeed, in the current state, comparison with other studies involving impaired subjects (e.g., in the discussion) has a questionable significance.

Other comments:

- Kinematics' results are reported in the Figures without the offset, but absolute values of data without removing the offset are comparing in the manuscript. How should these differences be interpreted? Comparing and presenting directly data without offset should be more straightforward.

- The standard deviation values of the RMS errors between experiments and simulation should be reported together with the mean values.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2023 Feb 10;18(2):e0270078. doi: 10.1371/journal.pone.0270078.r002

Author response to Decision Letter 0

6 Dec 2022

Please see our Response to Reviewers document for proper formatting of the text below.

We thank the editor and the reviewers for their thorough review of our work. Responding to the reviewers’ comments has improved this manuscript. The specific responses to the comments are described below, and the changes are highlighted in the revised manuscript. Please note that the sentences in italics are reviewer comments.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOSONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Authors’ Response: We have modified the manuscript to comply with style requirements for PLOS ONE.

2. Thank you for providing the following information in your Ethics Statement: "The Institutional Review Board at New Jersey Institute of Technology approved this study. The approval number is 2008001531R001." We ask that you also amend your Methods section to include this statement.

Authors’ Response: We added the requested sentence in the Methods section.

3. Thank you for stating in your Funding Statement:

“WAB and SP received support from the Department of Veterans Affairs (https://www.va.gov/) grant VA RR&D # 1 I01RX003561-01A2. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now. Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

Authors’ Response: We included an amended Funding Statement in our cover letter, as requested.

Reviewers' Comments:

5. Review Comments to the Author

Reviewer #1

This is a well-written paper that describes an experiment and simulator with one subject performing walking and sit-to-stand motions with and without an exoskeleton. I believe the manuscript is eventually worthy of publication, though I have a number of major and minor comments.

Authors’ Response: We thank the reviewer for their positive feedback.

MAJOR REVISION

1. The authors repeatedly emphasize that existing studies have not been performed on an FDA-approved exoskeleton, and that the novelty of the work is usage with an FDA-approved exoskeleton. However, it is unclear why the FDA approval represents a major difference. It seems like existing studies have achieved similar results with less popular exoskeletons, and most of the approaches used in existing studies can be approved to other exoskeletons with relatively little modification. The authors should better explain why the study is more than simply applying existing methods to a more popular exoskeleton.

Authors’ Response: We thank the reviewer for the opportunity to clarify why we restricted our work to an FDA-approved exoskeleton device. It is important to study the dynamics of the human-robot system during exoskeletal-assisted locomotion (EAL) in FDA-approved devices because individuals with neurological disorders only have access to FDA-approved devices. Currently, there is a disconnect between the exoskeletons used by patients and exoskeletons used to study human-robot interaction. Our study is a first step to address this disconnect. We have added several sentences in Introduction to highlight the importance of studying the dynamics of exoskeleton-assisted locomotion in FDA-approved devices.

Next, prior studies with FDA-approved exoskeletons have quantified joint kinematics, EMG, foot reaction forces, and ground reaction forces. To the best of our knowledge, no prior study has reported joint moments of the human-robot system during EAL in an FDA-approved device. We have clarified this gap in knowledge in Introduction section.

2. Related to the above: the Discussion acknowledges that several studies have already been done with the ReWalk exoskeleton (which the authors are using). The authors should better characterize how their work advances previous ReWalk work rather than simply saying their results are in line with previous ReWalk results.

Authors’ Response: We have added a paragraph in the Discussion section to highlight how our work advances previous work performed with the ReWalk exoskeleton, as requested.

3. Why did the authors recruit only a single subject for a single session? It seems like there would be significant intersubject variability that would both make it difficult to compare exoskeleton vs. no-exoskeleton conditions as well as make it difficult to evaluate performance of a simulation model. Additionally, since this was an able-bodied subject who only participated for a few sessions, it seems like the authors could have fairly easily obtained a larger sample size.

Authors’ Response: We agree with the Reviewer that there will be some inter-subject variability and additional participants will be required to validate the findings of this study. We respectfully disagree that adding additional participants to our study would be easy. To this point, conducting the controlled experiments and developing the subject-specific virtual simulators of walking, sit-to-stand, and stand-to-sit maneuvers with and without the exoskeleton are extremely laborious and time-consuming tasks. We currently are actively recruiting 7 able-bodied and 10 spinal cord injured participants to extend this work and anticipate that it will take at least a couple of years to recruit, study, and complete the analyses on these participants.

To clarify, the goal of this manuscript is to report our novel framework combining controlled experiments with computational modeling to build subject-specific virtual simulators of EAL that reproduces walking, sit-to-stand, and stand-to-sit maneuvers. Developing this framework represents substantial effort. We have described our methods in sufficient detail for the research community to reproduce our work. In addition, we will make all empirical data from this study freely available to the research community. There is benefit to publishing our dataset now so the modeling community can start building on our work to address a broad range of research questions. We have added several sentences in the section that addresses study limitations to clarify this point.

4. The simulation model does not appear to be extensively validated - the authors show that it can fairly accurately reproduce actual trajectories, but this is done by training and testing on the same subject in the same session. Is this sufficient for a real simulator? I would have expected it to be validated by examining predictive ability in different situations. And when the authors say that the simulator reproduced actual trajectories "within acceptable tolerances", what is an acceptable tolerance?

Authors’ Response: We agree that we have not yet tested the predictive capability of our virtual simulators. As we have stated previously in our response, the goal of this study was to develop the framework to build subject-specific virtual simulators of EAL. Testing the predictive capability of these virtual simulators will be a component of our future work. We invite other investigators to use our dataset to test the predictive capability of our virtual simulators, or to build their own virtual simulators from our empirical data. We have clarified in the Abstract that this work provides a foundation for parametric studies to characterize the effects of human and robot design variables, and predictive modeling to optimize human-robot interaction during EAL. We have also added a few sentences at the end of the Discussion section encouraging other investigators to build on our work.

Next, acceptable tolerance is less than 2 cm average RMS error between experiment and simulator markers, per OpenSim’s best practices, which has been added with a sentence and two references in the Results section.

MINOR REVISION

1. Is it really necessary to list inclusion/exclusion criteria when only a single participant was recruited? Authors list things like pregnancy and lactation being exclusion criteria, which seems irrelevant for a male participant. They also list minimum and maximum weight, which again seems irrelevant since a single participant was recruited and his weight is known.

Authors’ Response: We thank the reviewer for this comment and in response have removed the inclusion/exclusion criteria from the manuscript.

2. Could the authors briefly describe how maximum voluntary contraction protocols were performed?

Authors’ Response: We have added several sentences in the Methods section to describe our maximum voluntary contraction trials.

3. Why were there 4 trials of exoskeleton-assisted sit-to-stand and 5 trials of exoskeleton-unassisted sit-to-stand? Why not 5 for both?

Authors’ Response: We collected 10 trials each of walking, sit-to-stand, and stand-to-sit maneuvers, but only successful trials were included for further analyses. We have clarified in the Methods section the criteria for a successful trial for each maneuver. Meeting these criteria resulted in different numbers of trials for the different maneuvers. Based on these criteria, we obtained 6 successful trials each of exoskeletal-assisted and unassisted walking, 4 and 5 successful trials of exoskeletal-assisted and unassisted sit-to-stand maneuvers, respectively, and 5 successful trials each of exoskeletal-assisted and unassisted stand-to-sit maneuvers.

Reviewer #2

The manuscript presents a dataset collected on a healthy subject performing exoskeleton-assisted vs unassisted locomotion using an FDA-approved exoskeleton (ReWalk P6.0), and a subject-specific virtual simulator implemented using OpenSim environment, to compute joint kinematics (from motion capture and simulation) and moments (from simulation) for hip, knee and ankle in assisted vs unassisted conditions. In addition, EMG data from 8 lower-limb muscles were recorded, as well as ground reaction forces using force plates.

The main novelty claimed by the authors is the availability of such a dataset and simulation that includes an FDA-approved exoskeleton comparing assisted and unassisted locomotion. Nevertheless, whereas the need for having such a dataset and simulation is well explained in the manuscript, it is not clear which is the added value of including an FDA-approved exoskeleton, with respect to other examples (cited by the authors in the manuscript) using different exoskeletons. In my opinion, this alone cannot be the main claim of the manuscript given that: no novel methods are presented; different exoskeletons might result in different human-robot interaction when assistance is delivered; other similar data in literature could be used for the simulator if the benefit would be the possibility to "conduct rapid design-phase evaluations to accelerate device refinements" (Discussion).

Authors’ Response: We thank the reviewer for this comment and have addressed the importance of studying the dynamics of exoskeleton-assisted locomotion in FDA-approved devices in Reviewer 1’s first major revision comment. To briefly reiterate our prior response, it is important to study the dynamics of the human-robot system during EAL in FDA-approved devices because individuals with neurological disorders only have access to FDA-approved devices. Currently, there is a disconnect between the exoskeletons used by patients and exoskeletons used to study human-robot interaction. Our study is a first step to address this disconnect. We have added several sentences in the Introduction section to highlight the importance of studying the dynamics of exoskeleton-assisted locomotion in FDA-approved devices.

Next, our framework that combines controlled experiments with computational modeling to build subject-specific virtual simulators of EAL is novel. No prior study has simulated EAL in any FDA-approved exoskeleton. The minimum empirical dataset we will share as part of this publication is novel. There is no such dataset currently available to the research community, which explains why the computational modeling community has developed virtual simulators based on idealized or imaginary exoskeletons (Bianco et al., 2022; Dembia et al., 2017; Khamar et al., 2019; Nguyen et al., 2019; Uchida et al., 2016; Zhu et al., 2015). Furthermore, we have added two paragraphs in the Discussion section to highlight how our work advances previous work performed with the ReWalk and other FDA-approved exoskeletons.

The authors have stated clearly the limitations of the study, including having no participants with altered gait patterns (e.g., spinal cord) and not separating human and robotic dynamics in the virtual simulator (one aspect that was described in the Introduction as particularly important and not very well explored in the state of the art). Exploring further more at least one of these limitations would be important to justify the significance of the manuscript. Indeed, in the current state, comparison with other studies involving impaired subjects (e.g., in the discussion) has a questionable significance.

Authors’ Response: We thank the reviewer for this comment and have modified our text to provide a more detailed discussion of the limitation of including a single able-bodied participant in this study. Next, we have modified the text in the Discussion section to clarify how our work advances previous work done with the ReWalk and other FDA-approved exoskeletons.

Other comments:

- Kinematics' results are reported in the Figures without the offset, but absolute values of data without removing the offset are comparing in the manuscript. How should these differences be interpreted? Comparing and presenting directly data without offset should be more straightforward.

Authors’ Response: This comment is somewhat unclear to the authors. However, in our attempt to clarify, our segment coordinate frames were obtained with reference to a single global coordinate frame. The initial values of the joint kinematics from the different segment coordinate frames were based on this single global coordinate frame. This provided a direct comparison of joint kinematics with and without the exoskeleton. For example, using this approach we clearly see that the participant’s hips were more flexed during exoskeletal-assisted walking compared to unassisted walking (Figs 2A and 2B). If we were to remove the offset angles between the different segment coordinate axes, this would mask many of the kinematic differences that are of interest when analyzing locomotion with and without the exoskeleton. In our results, we have only subtracted the offset angles to calculate the average absolute differences in kinematics between the human and the robot during exoskeletal-assisted walking (Figs 2A-2D). To clarify the rationale for the approach taken, text in the Results section has been appropriately edited.

- The standard deviation values of the RMS errors between experiments and simulation should be reported together with the mean values.

Authors’ Response: We have included the standard deviation values of the RMS errors between experiment and simulator markers, as requested.

Attachment

Submitted filename: Response to Reviewers.pdf

Click here for additional data file. (188.5KB, pdf)

Decision Letter 1

Imre Cikajlo

8 Jan 2023

PONE-D-22-15991R1Comparison of the dynamics of exoskeletal-assisted and unassisted locomotion in an FDA-approved lower extremity device: Controlled experiments and development of a subject-specific virtual simulatorPLOS ONE

Dear Dr. Pal,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

​Please submit your revised manuscript by Feb 22 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Imre Cikajlo, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

Please pay attention to the issues raised by the reviewer 1.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have partially addressed my comments from the first round of review, and I believe the paper is now somewhat closer to publication. However, I have several remaining comments and thus recommend another round of major revisions.

MAJOR ISSUES

1. The authors repeatedly emphasize that the novel dataset collected during the study would be of value to researchers. However, as far as I can tell, they have not made the dataset available, and present conflicting statements in the PLOS ONE submission system. The Data Availability box states "Yes - all data are fully available without restriction", but the "Describe where the data may be found in full sentences" box then simply states "All empirical data from this study will be freely available." The Discussion also states that all data "will be" fully available. If the novel dataset is part of the value of the manuscript, the authors should make it available during the review process. If that is not possible for some exceptional reason, the authors should at least provide some indication of how the data will be made available so that the approach can be evaluated.

2. The authors state that, to their knowledge, no previous study has reported joint moments. I would appreciate a brief explanation of how difficult it is to obtain these joint moments and what the usefulness of joint moments is relative to other data that was already previously done. This would help evaluate the degree of methodological novelty.

3. I agree that patients have access only to FDA-approved devices, but at the same time, the authors claim that the goal of their work is to "report our novel framework combining controlled experiments with computational modeling to build subject-specific virtual simulators". A novel framework should be applicable to many exoskeletons, not just the specific FDA-approved one, and it is thus still not convincing why the authors emphasize FDA approval so much. Is the novelty that this is the first empirical dataset of this scope with an FDA-approved device?

4. The authors' justification for why only 1 subject is possible is that it would take years to conduct a study with ~20 participants. If development of virtual simulators for specific individuals is so time-consuming (seemingly on the scale of months / participant), the authors should better justify the expected practical value of such simulators. For real patients, economic and logistical issues would likely not allow months of time to be devoted to tuning a simulator to each person.

MINOR ISSUES

1. The authors' justification for why it is important to study the dynamics of the human-robot system during EAL in FDA-approved devices appears to have been awkwardly dropped into the Introduction without much effort to make it "flow" well with the rest of the text. Would recommend better integrating with previously existing text.

2. Lines 171-172: "A generic musculoskeletal [55] was scaled to the participant’s anthropometry." A noun seems to be missing here.

3. The authors state that they conducted 10 trials for each activity and then analyzed only the successful trials, leading to different numbers of trials in each activity. Why did the authors alternatively not repeat trials until a certain number of successful trials was obtained? Is it not possible to check for missing data within the session?

4. The Introduction says that the new dataset is novel because it includes joint moments, which had not been previously available. The Discussion then says that the new dataset is novel because it includes joint moments and EMG. If the EMG is also novel, I recommend mentioning that in the Introduction as well.

Reviewer #2: The authors have addressed the Reviewers' comments and the contribution of the manuscript is more clear.

I would suggest to carefully check the new text that was included to avoid repetitions (i.e., the sentence "we will make all empirical data from this study freely available to the research community" is repeated twice in the Discussion). Regarding this point, it would be more sound to write in the manuscript that the dataset is already available (instead of "will be available")

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Decision Letter 2

Imre Cikajlo

25 Jan 2023

Comparison of the dynamics of exoskeletal-assisted and unassisted locomotion in an FDA-approved lower extremity device: Controlled experiments and development of a subject-specific virtual simulator

PONE-D-22-15991R2

Dear Dr. Pal,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Imre Cikajlo, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have adequately addressed my comments, and I am fine with the paper being accepted as-is. I would recommend discretionary changes:

Consider making changes to the text in response to my last major issue #4 and minor issue #3. While the authors' response (in the Response to Reviewer document) is adequate, the authors do not appear to have changed the text in response, and other readers may have the same questions. For reference, the previously raised issues were:

- The authors' justification for why only 1 subject is possible is that it would take years to conduct

a study with ~20 participants. If development of virtual simulators for specific individuals is so

time-consuming (seemingly on the scale of months / participant), the authors should better justify

the expected practical value of such simulators. For real patients, economic and logistical issues

would likely not allow months of time to be devoted to tuning a simulator to each person.

- The authors state that they conducted 10 trials for each activity and then analyzed only the

successful trials, leading to different numbers of trials in each activity. Why did the authors

alternatively not repeat trials until a certain number of successful trials was obtained? Is it not

possible to check for missing data within the session?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

**********

Acceptance letter

Imre Cikajlo

2 Feb 2023

PONE-D-22-15991R2

Comparison of the dynamics of exoskeletal-assisted and unassisted locomotion in an FDA-approved lower extremity device: Controlled experiments and development of a subject-specific virtual simulator

Dear Dr. Pal:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Imre Cikajlo

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Fig. Average (±1 SD) joint angles from four exoskeletal-assited (A-F; human: blue for left leg, red for right leg; robot: black) and five unassisted (G-L; human: blue for left leg, red for right leg) sit-to-stand trials.

    The offset in angles between the human (blue or red) and the robot (black) was due to different definitions of coordinate axes for each rigid body. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver.

    (TIF)

    Click here for additional data file. (480.5KB, tif)
    S2 Fig. Average (±1 SD) ground reaction forces from four exoskeletal-assited (A-C) and five unassisted (D-F) sit-to-stand trials.

    The ground reaction forces from exoskeletal-assisted locomotion were normalized to the combined weight of the participant and the exoskeleton; forces from unassisted locomotion were normalized to the participant’s body weight (BW). (A-C) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver. A-P = Anterior-Posterior, M-L = Medial-Lateral.

    (TIF)

    Click here for additional data file. (173.1KB, tif)
    S3 Fig. Average (+1 SD) normalized electromyography (EMG) data from four exoskeletal-assisted (A-H) and five unassisted (I-P) sit-to-stand trials.

    The EMG data were normalized using muscle-specific maximum voluntary contraction values. (A-H) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver. RF = rectus femoris, VL = vastus lateralis, VM = vastus medialis, ST = semitendinosus, BF = biceps femoris, GM = gastrocnemius medialis, SOL = soleus, TIB = tibialis anterior.

    (TIF)

    Click here for additional data file. (180.1KB, tif)
    S4 Fig. Average (±1 SD) joint moments from four exoskeletal-assited (A-F) and five unassisted (G-L) sit-to-stand trials.

    The joint moments from exoskeletal-assisted locomotion were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by flexing the torso forward and loading the crutches) to hold phase (leaned forward for 3 seconds). The second dashed vertical lines represent the beginning of the rise phase, with the hip and knee motors activating to complete the sit-to-stand maneuver.

    (TIF)

    Click here for additional data file. (185.4KB, tif)
    S5 Fig. Average (±1 SD) joint angles from five exoskeletal-assited (A-F; human: blue for left leg, red for right leg; robot: black) and five unassisted (G-L; human: blue for left leg, red for right leg) stand-to-sit trials.

    The offset in angles between the human (blue or red) and the robot (black) was due to different definitions of coordinate axes for each rigid body. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver.

    (TIF)

    Click here for additional data file. (489.8KB, tif)
    S6 Fig. Average (±1 SD) ground reaction forces from five exoskeletal-assited (A-C) and five unassisted (D-F) stand-to-sit trials.

    The ground reaction forces from exoskeletal-assisted locomotion were normalized to the combined weight of the participant and the exoskeleton; forces from unassisted locomotion were normalized to the participant’s body weight (BW). (A-C) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver. A-P = Anterior-Posterior, M-L = Medial-Lateral.

    (TIF)

    Click here for additional data file. (161.5KB, tif)
    S7 Fig. Average (+1 SD) normalized electromyography (EMG) data from five exoskeletal-assisted (A-H) and five unassisted (I-P) stand-to-sit trials.

    The EMG data were normalized using muscle-specific maximum voluntary contraction values. (A-H) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver. RF = rectus femoris, VL = vastus lateralis, VM = vastus medialis, ST = semitendinosus, BF = biceps femoris, GM = gastrocnemius medialis, SOL = soleus, TIB = tibialis anterior.

    (TIF)

    Click here for additional data file. (161.9KB, tif)
    S8 Fig. Average (±1 SD) joint moments from five exoskeletal-assited (A-F) and five unassisted (G-L) stand-to-sit trials.

    The joint moments from exoskeletal-assisted locomotion were normalized to the combined mass of the participant and the exoskeleton; joint moments from unassisted locomotion were normalized to the participant’s mass. (A-F) The first dashed vertical lines represent the transition from positioning phase (user gets in to position by leaning backwards and loading the crutches) to hold phase (leaned backward for 3 seconds). The second dashed vertical lines represent the beginning of the descend phase, with the hip and knee motors activating to complete the stand-to-sit maneuver.

    (TIF)

    Click here for additional data file. (165.1KB, tif)
    Attachment

    Submitted filename: Response to Reviewers.pdf

    Click here for additional data file. (188.5KB, pdf)
    Attachment

    Submitted filename: Response to Reviewers.pdf

    Click here for additional data file. (170.5KB, pdf)

    Data Availability Statement

    All relevant data are available on Figshare: https://doi.org/10.6084/m9.figshare.21967874.v1.

    Articles from PLOS ONE are provided here courtesy of PLOS

    RESOURCES