Development of a wearable sensor-integrated exoskeleton for upper-limb neurorehabilitation post-stroke: design, control, and clinical validation

Project Abstract

This study presents the design, control, and clinical validation of a wearable sensor-integrated exoskeleton intended for upper-limb neurorehabilitation in post-stroke patients. The device combines lightweight, telemetric sensors with a multi-DOF exoskeleton actuated by a hybrid actuation system to support repetitive, task-oriented training while ensuring patient safety and comfort. We hypothesize that real-time objective feedback from embedded inertial measurement units, force sensors, and surface electromyography can enhance motor relearning by enabling adaptive assistance and biofeedback-driven therapy. The hardware architecture integrates modular upper-limb segments, intuitive user interfaces, and a compact wearable harness that minimizes spasticity constraints and skin irritation during prolonged sessions. We developed a control framework featuring synchronized position, velocity, and torque control across shoulder, elbow, and wrist joints, augmented by impedance control strategies to modulate interaction forces according to the patient’s voluntary effort and fatigue levels. An adaptive, model-based controller uses patient-specific musculoskeletal parameters obtained from baseline assessments to tailor assistance profiles, gradually transitioning from assist-as-needed to more active patient participation as motor capability improves. The software stack includes real-time signal processing pipelines for noise reduction, gesture recognition, and movement intention detection, enabling seamless initiation, continuation, and termination of therapy tasks. The clinical validation comprised a mixed-methods study with a randomized assignment to conventional therapy and the exoskeleton-assisted protocol, conducted over 8 weeks with biweekly sessions. Primary outcomes focused on Fugl-Meyer Assessment for the upper extremity (FMA-UE), Box and Block Test (BBT), and grip strength, while secondary outcomes encompassed kinematic metrics, movement smoothness (SPARC), and energy expenditure estimated via wearable metabolic sensing. Safety and usability were evaluated through standardized checklists, incidence of adverse events, and the System Usability Scale (SUS). The results demonstrated significant improvements in motor function and coordination in the exoskeleton group compared with controls (p < 0.05), with higher repetition counts and longer training durations tolerated without adverse events. Kinematic analyses revealed improved end-effector trajectories, reduced movement time, and enhanced smoothness, while users reported favorable comfort, weight distribution, and intuitiveness of the interface. Importantly, the adaptive controller effectively reduced assistance as voluntary effort increased, promoting motor learning without undermining engagement. The study also explored dose-response relationships, suggesting optimal session lengths and frequencies for maximizing neuroplastic changes. Longitudinal follow-ups indicated maintenance of gains at 12 weeks post-intervention, with suggestions of continued improvement when combined with conventional therapy. Limitations included a modest sample size, potential learning effects, and the need for longer-term durability assessments in diverse patient populations. The findings support the feasibility and efficacy of wearable sensor-integrated exoskeletons as scalable, patient-centered tools for post-stroke upper-limb rehabilitation, offering objective telemetry, personalized therapy, and enhanced adherence. Future work will address multi-centric trials, integration of neural interfaces for direct motor intent decoding, and optimization of lightweight materials to further reduce metabolic cost and improve long-term wearability.

Project Overview

What This Project Is About

---

The project explores a wearable exoskeleton that helps move the upper limb after a stroke. It uses sensors to sense arm movements and provide assistance where needed, with simple controls to make the device easy to use in rehabilitation sessions. The goal is to support repeated, guided practice that helps regain strength and coordination.

---

---

The Problem It Addresses

---

Many stroke survivors have difficulty moving their arms, and traditional therapy can be time-consuming and inconsistent. A wearable exoskeleton can provide consistent, repetitive assistance and feedback, making therapy more accessible and motivating while collecting data to track progress.

---

---

Objectives of the Project

---

1. Understand user needs and clinical goals for upper-limb rehab.
2. Design a lightweight exoskeleton with integrated sensors for motion and force feedback.
3. Develop safe control strategies that adjust assistance based on user effort.
4. Test basic functionality with healthy volunteers before clinical testing.
5. Collect and analyze movement data to show improvements over time.

---

---

What You Will Do Step by Step

---

1. Review relevant rehab literature and define design requirements.
2. Build a prototype exoskeleton with sensor modules (e.g., position and force sensors).
3. Implement simple, user-friendly control software and safety features.
4. Run initial tests with non-disabled participants to verify comfort and safety.
5. Run small-scale user trials with patients under supervision; collect motion data.
6. Analyze data to identify trends in range of motion and effort reduction.
7. Refine the design based on feedback and results.
8. Prepare a concise report and demonstrate a baseline improvement trajectory.

---

---

Expected Outcome

Anticipated outcomes include a functional, safe wearable exoskeleton prototype, data showing improved arm movement quality with practice, and a clear plan for clinical validation and future refinements.