Learning Therapist Policy from Therapist-Exoskeleton-Patient Interaction
Authors: Grayson Snyder, Lorenzo Vianello, Levi Hargrove, Matthew L. Elwin, Jose Pons
Organizations: Center for Robotics and Biosystems, Northwestern University, Evanston, IL, USA. · Legs and Walking Lab, Shirley Ryan AbilityLab, Chicago, IL, USA.
Post-stroke rehabilitation is often necessary for patients to regain proper walking gait. However, the typical therapy process can be exhausting and physically demanding for therapists, potentially reducing therapy intensity, duration, and consistency over time. We propose a Patient-Therapist Force Field (PTFF) to visualize therapist responses to patient kinematics and a Synthetic Therapist (ST) machine learning model to support the therapist in dyadic robot-mediated physical interaction therapy. The first encodes patient and therapist stride kinematics into a shared low-dimensional latent manifold using a Variational Autoencoder (VAE) and models their interaction through a Gaussian Mixture Model (GMM), which learns a probabilistic vector field mapping patient latent states to therapist responses. This representation visualizes patient-therapist interaction dynamics to inform therapy strategies and robot controller design. The latter is implemented as a Long Short-Term Memory (LSTM) network trained on patient-therapist interaction data to predict therapist-applied joint torques from patient kinematics. Trained and validated using leave-one-out cross-validation across eight post-stroke patients, the model was integrated into a ROS-based exoskeleton controller to generate real-time torque assistance based on predicted therapist responses. Offline results and preliminary testing indicate the potential of their use as an alternative approach to post-stroke exoskeleton therapy. The PTFF provides understanding of the therapist's actions while the ST frees the human therapist from the exoskeleton, allowing them to continuously monitor the patient's nuanced condition.
Figures & tables
Fig. 1: Learning Therapist Policy from Therapist–Exoskeleton–Patient Interaction. (A) Previously collected dataset of therapist–exoskeleton–patient physical interaction through two lower-limb exoskeletons. The exoskeletons render virtual springs connecting the joint configurations of the patient and therapist. (B) Patient–Therapist Force Field, which takes as input the patient’s and therapist’s strides, extracts latent representations, and learns a force field connecting them in a lower-dimensional space. The model uses an encoder–decoder architecture (both displayed) to reduce dimensionality. (C) Synthetic Therapist, which takes as input the patient’s joint kinematics and predicts the corresponding therapist kinematics. The model uses a Long Short-Term Memory (LSTM) network (displayed) to infer the therapist’s trajectory.
Fig. 2: Top: Latent space representation of each individual dyad. Each dot corresponds to a stride, while the ellipsoids display their distribution. Blue: patient left stride; green: therapist right stride; light blue: patient right stride; light green: therapist left stride. The connection is mirrored (patient left, therapist right). In the middle the resulting force field computed by fitting all the dyads using the GMM. Bottom: Force fields (from left to right): patient hip, therapist hip, patient knee, therapist knee.
rMSE (deg, deg/s)
Train/Validation-Set
Test-Set
Position Avg.
Velocity Avg.
PT 1-8
PT 1-8
4.30±0.41
20.67±1.47
PT 2-8
PT 1
5.93±1.20
25.13±4.54
PT 1,3-8
PT 2
5.77±1.12
25.47±5.03
PT 1,2,4-8
PT 3
6.48±0.67
33.62±3.87
PT 1-3,5-8
PT 4
4.95±1.00
20.66±1.70
TABLE I: Prediction accuracy expressed as mean root Mean-Squared-Error (rMSE) for the LSTM model joint position and joint velocity prediction. We performed a cross-user generalization evaluation by iteratively training on all users (PT i ) except one and testing on the excluded user.
Fig. 3: Example of Synthetic Therapist predictions (predicting 75ms into the future) overlayed with corresponding true therapist data and patient data. On top is presented the Hip Joint, while on the bottom the Knee joint.
Upper extremity motor function recovery is positively linked to Task-Specific Training (TST) and sufficient therapy dosage. Rehabilitation robots can increase TST dosage via controlled, repetitive treatment and free therapists to simultaneously manage other patients, but it has yet to demonstrate significant benefits over conventional treatment. This is potentially linked to inaccurate robotic representation of personalised physical therapist-patient interaction and lack of practice variability during TST. Hence, we advocate for robotic interventions that preserve the personalised physical therapist-patient interactions when delivering TST for patients across varying practise conditions. We propose a Learning-from-Demonstration framework using Task-Parameterised Gaussian Mixture Models (TPGMM) to learn personalised physical therapist-patient interaction in Task-Specific exercises, mapping patient joint kinematics to therapist-applied torques using few demonstrations. The model is generalised to reconstruct therapist torques in new task variations. The framework was evaluated on physical interactions from 14 mock "therapist-patient" pairs over three tasks of increasing complexity, each with six variations. A benchmark comparison against a Look-Up Table was conducted. The results show both methods reproducing interactions in unseen task variations that deviate slightly from the actual interaction, with TPGMM slightly outperforming LUT. Both methods reproduced interactions that gets increasingly closer to the actual interaction as task complexity increases.
Jia Quan Loh, Vincent Crocher, Marlena Klaic +2
Human Robotics Laboratory, Department of Mechanical Engineering, The University of Melbourne · Melbourne School of Health Sciences, The University of Melbourne
Robotic systems can enhance the amount and repeatability of physically guided motor training. Yet their real-world adoption is limited, partly due to non-intuitive trainer/therapist-trainee/patient interactions. To address this gap, we present a haptic teleoperation system for trainers to remotely guide and monitor the movements of a trainee wearing an arm exoskeleton. The trainer can physically interact with the exoskeleton through a commercial handheld haptic device via virtual contact points at the exoskeleton's elbow and wrist, allowing intuitive guidance. Thirty-two participants tested the system in a trainer-trainee paradigm, comparing our haptic demonstration system with conventional visual demonstration in guiding trainees in executing arm poses. Quantitative analyses showed that haptic demonstration significantly reduced movement completion time and improved smoothness, while speech analysis using large language models for automated transcription and categorization of verbal commands revealed fewer verbal instructions. The haptic demonstration did not result in higher reported mental and physical effort by trainers compared to the visual demonstration, while trainers reported greater competence and trainees lower physical demand. These findings support the feasibility of our proposed interface for effective remote human-robot physical interaction. Future work should assess its usability and efficacy for clinical populations in restoring clinicians' sense of agency during robot-assisted therapy.
Beatrice Luciani, Alex van den Berg, Matti Lang +2
Department of Cognitive Robotics, Delft University of Technology, Delft, The Netherlands · Department of Rehabilitation Medicine, Erasmus Medical Center, Rotterdam, The Netherlands.
Successful robot-mediated rehabilitation requires designing games and robot interventions that promote healthy motor practice. However, the interplay between a given user's neuromotor behavior, the gaming interface, and the physical robot makes designing system elements -- and even characterizing what behaviors are "healthy" or pathological -- challenging. We leverage our OpenRobotRehab 1.0 open access data set to assess the characteristics of 13 healthy and 2 post-stroke users' force output, muscle activations, and game performance while executing isometric trajectory tracking tasks using an end-effector rehabilitation robot. We present an assessment of how subtle aspects of interface design impact user behavior; an analysis of how pathological neuromotor behaviors are reflected in end-effector force dynamics; and a novel hidden Markov model (HMM)-based neuromotor behavior classification method based on surface electromyography (sEMG) signals during cyclic motions. We demonstrate that task specification (including which axes are constrained and how users interpret tracking instructions) shapes user behavior; that pathology-related features are detectable in 6D end-effector force data during isometric task execution (with significant differences between healthy and post-stroke profiles in force error and average force production at p=0.05); and that healthy neuromotor strategies are heterogeneous and inherently difficult to characterize. We also show that our HMM-based models discriminate healthy and post-stroke neuromotor dynamics where synergy-based decompositions reflect no such differentiation. Lastly, we discuss these results' implications for the design of adaptive end-effector rehabilitation robots capable of promoting healthier movement strategies across diverse user populations.
Ajay Anand, Gabriel Parra, Chad A. Berghoff +1
The authors are with the Departments of Mechanical Engineering, Physical Medicine & Rehabilitation, and Mathematics, the Kahlert School of Computing, and the Robotics Center at the University of Utah, Salt Lake City, UT 84112, USA.