Adaptive Hip Exoskeleton for Stroke Survivors With Gait Impairment
Overview
- Phase
- Not Applicable
- Intervention
- Not specified
- Conditions
- Stroke
- Sponsor
- Georgia Institute of Technology
- Enrollment
- 12
- Locations
- 1
- Primary Endpoint
- Temporal Convolutional Network (TCN) model performance (Joint moment accuracy)
- Status
- Recruiting
- Last Updated
- 6 months ago
Overview
Brief Summary
This work will focus on new algorithms for robotic exoskeletons and testing these in human subject tests. Individuals who have previously had a stroke will walk while wearing a robotic exoskeleton on a specialized treadmill as well as during other movement tasks (e.g. over ground, stairs, ramps). The study will compare the performance of the advanced algorithm with not using the device to determine the clinical benefit.
Detailed Description
The focus of this work is a proposed novel artificial intelligence (AI) system to self-adapt control policy in powered exoskeletons to aid deployment systems that personalize to individual patient gait. Individuals post stroke have a broad range of mobility challenges including asymmetric gait, substantially decreased SSWS, and reduced stability, and therefore have greatly impaired overall mobility independence in the community. The investigators expect the proposed novel controller, capable of personalization to such variable and asymmetric gait patterns, will have significant benefits towards increasing community independence and mobility for patients post stroke. Patients post stroke will be fit with a hip exoskeleton (in a powered and/or unpowered state) and proceed to walk on a treadmill or perform various movement tasks. The same tasks will be performed by the patients without wearing the hip exoskeleton to serve as a baseline. The investigators expect improved outcomes in the powered hip exoskeleton compared to the unpowered hip exoskeleton and baseline conditions.
Investigators
Eligibility Criteria
Inclusion Criteria
- •Between 18-85 years of age
- •Had a stroke at least 6 months prior to study involvement
- •Are community dwelling, which means the participant does not live in an assisted living facility
- •Are able to provide informed consent to participate in the study activities
- •Can safely participate in the study activities (per self-report)
- •Must have a Functional Ambulation Category (FAC) score of 3 or above, which means the participant can walk without the assistance of another person
Exclusion Criteria
- •Require a walker to walk independently
- •Have a shuffling gait pattern overground
- •Have a Functional Ambulation Category (FAC) score of 2 or lower, which means the participant requires the assistance of another person in order to walk
- •Have a significant secondary deficit beyond stroke (e.g. amputation, legal blindness or other severe impairment or condition) that in the opinion of the Principal Investigator (PI), would likely affect the study outcome or confound the results
- •For exoskeleton-only studies, the exoskeleton device does not fit appropriately or safely, as determined by the research team during the fitting assessment.
Outcomes
Primary Outcomes
Temporal Convolutional Network (TCN) model performance (Joint moment accuracy)
Time Frame: 1 year
This outcome represents the error with which the deep learning model embedded into our hip exoskeleton's microprocessor predicts hip joint moments in stroke patients. Specifically, the coefficient of determination (R²) is computed between the predicted hip joint moments and the ground truth measurements. Ground truth measurements are obtained from a laboratory-grade force plate system and inverse dynamics calculations. Hip joint moment predictions are made at a frequency of 200 Hz and compared to the laboratory-measured values. For these measures, higher R² values (closer to 1.0) indicate better correlation between predicted and actual hip joint moments. This metric provides a comprehensive assessment of the exoskeleton's ability to accurately estimate hip joint moments in stroke patients during tasks, with improved outcomes representing better assistive capabilities for the user.
Metabolic cost for level ground walking
Time Frame: 1 year
Metabolic energy expenditure will be quantified using an indirect calorimetry system (Parvo Medics, UT) that measures oxygen consumption (VO₂) and carbon dioxide production (VCO₂) during experimental tasks. Measurements will be collected from each participant during a 5-minute baseline standing period followed by level ground walking trials under three conditions: without the exoskeleton, with the exoskeleton in a powered state, and with the exoskeleton in an unpowered state. Metabolic cost will be calculated from respiratory gas exchange data using standard equations for energy expenditure.
Biological joint work
Time Frame: 1 year
Mechanical work performed by the lower limb joints will be quantified through biomechanical analysis of motion capture data. Joint moments and angular velocities will be derived through inverse dynamics and kinematics, respectively. Joint power, calculated as the product of joint moment and angular velocity, will be integrated with respect to time using trapezoidal integration to determine mechanical work. Positive and negative work will be calculated by separately integrating positive and negative joint powers, providing comprehensive quantification of joint energy generation and absorption at each joint during the movement tasks.
Secondary Outcomes
- Single limb stance time asymmetry index(1 year)
- Step Length Asymmetry index(1 year)
- 10 meter walk test (self-selected)(1 year)
- The timed up and go (TUG)(1 year)
- 6 Minute Walk Test(1 year)
- Modified Stroke Impact Scale(1 year)
- Modified Activities-specific balance confidence(1 year)
- Fast self-selected walking speed(1 year)