Assistive Hip Exoskeleton Study for Stroke

Improving Community Ambulation for Stroke Survivors Using Powered Hip Exoskeletons With Adaptive Environmental Controllers

The increased metabolic and biomechanical demands of ambulation limit community mobility in persons with lower limb disability due to neurological damage. There is a critical need for improving the locomotion capabilities of individuals with stroke to increase their community mobility, independence, and health. Robotic exoskeletons have the potential to assist these individuals by increasing community mobility to improve quality of life. While these devices have incredible potential, current technology does not support dynamic movements common with locomotion such as transitioning between different gaits and supporting a wide variety of walking speeds. One significant challenge in achieving community ambulation with exoskeletons is providing an adaptive control system to accomplish a wide variety of locomotor tasks. Many exoskeletons today are developed without a detailed understanding of the effect of the device on the human musculoskeletal system. This research is interested in studying the question of how the control system affects stroke biomechanics including kinematic, kinetics and muscle activation patterns. By optimizing exoskeleton controllers based on human biomechanics and adapting control based on task, the biggest benefit to patient populations will be achieved to help advance the state-of-the-art with assistive hip exoskeletons.

One significant challenge in achieving community ambulation with exoskeletons is providing an adaptive control system to accomplish a wide variety of locomotor tasks. Many exoskeletons today are developed without a detailed understanding of the effect of the device on the human musculoskeletal system. The study is interested in exploring the question of how the control system affects human biomechanics including kinematic, kinetics and muscle activation patterns. By optimizing exoskeleton controllers based on human biomechanics and adapting control based on task, this work will be able to provide the biggest benefit to patients and advance the state-of-the-art with assistive hip exoskeletons. A large patient population that could benefit from lower limb assistive technology are stroke survivors, which is the specific population this proposal targets. One common characteristic of stroke survivors who regain their ability to walk is that the hip muscles are overtaxed due to distal weakness. The investigators propose to use a powered hip exoskeleton to augment their proximal musculature, which needs to produce significant power output in most locomotion activities such as standing up, walking, and going up stairs or slopes. Another biomechanical aspect of stroke survivors is an asymmetric gait in terms of kinematics, kinetics and muscle activations. The research will examine what kind of exoskeleton assistance is most beneficial to stroke survivors for enhancing community ambulation. The hypothesis is that since the gait is asymmetric, the controller will need to be asymmetric to provide optimal assistance to aid in mobility. The long-term research goal is to create powered assistive exoskeletons devices that are of great value to individuals with serious lower limb disabilities by improving clinical outcomes such as walking speed and community ambulation ability. The overall objective of the proposed project is to study the biomechanical effects of using a hip exoskeleton with adaptive controllers for assisting stroke survivors with lower limb deficits to improve their community ambulation capabilities. The central hypothesis overarching both aims is that exoskeleton control that adapts to environmental terrain will improve mobility metrics for human exoskeleton users on community ambulation tasks. The rationale is that since human biomechanics change based on task, exoskeleton controllers likewise need to optimize their assistance levels to match what the human is doing. The team has previously designed and extensively tested an autonomous hip exoskeleton in able-bodied subjects on a treadmill and plan to follow this up with a separate study on able bodied subjects during overground locomotion of walking, stairs, and ramps. The aim of this study is to translate an autonomous robotic hip exoskeleton to provide adaptive assistance in community ambulation for stroke survivors with mobility impairment. The team will analyze the biomechanical effects and clinical benefits with using an autonomous hip exoskeleton for a walking impaired user (due to stroke). The primary hypothesis for this aim is that stroke survivors will increase their mobility in community ambulation tasks using the adaptive control framework. A sub-hypothesis is that stroke survivors who present with unilateral impairment will have superior biomechanical and clinical outcomes using a controller with asymmetric assistance. The investigators expect a controller that provides a greater assistance to the impaired side to improve overall symmetry and help the stroke survivor maintain a more efficient gait pattern to help improve walking speed (primary outcome measure). The expected outcome of these aims will be an increased understanding of the biomechanical and clinical effects in applying hip assistance with a robotic exoskeleton in community ambulation tasks such as overground walking, ramps and stairs. This work will serve as a foundational start for a broader planned study of optimizing controllers to improve biomechanics in the walking impaired using powered hip autonomous exoskeletons. This aim will have a positive impact by helping to inform the design and control of future exoskeleton for assisting individuals with lower limb disabilities, with specific insight in stroke survivors with mobility impairment.

The model used is a repeated measures single arm study. Multiple conditions including using and not using the device will be tested on the same subjects to have multiple test points on a per subject basis.

This study will be conducted on a sample population of stroke subjects (single arm). Each subject will test with each condition of the exoskeleton (repeated measures).

The study team will be testing a powered hip exoskeleton and its capability to improve locomotion in stroke survivors.

Using five different hip exoskeleton assistance strategies, the participant's overground self-selected walking speed was recorded. Assistance types are 1) Unilateral Paretic Assistance, 2) Unilateral Non-Paretic Assistance, 3) Bilateral Equal Assistance, 4) Bilateral Additional Paretic Assistance, and 5) Bilateral Additional Non-Paretic Assistance. The first information (unilateral or bilateral) refers to the leg(s) that the exoskeleton is providing assistance with. For example, unilateral assistance means that the assistance is provided to only one side (zero assistance for the other side). The second information (additional paretic/non-paretic or equal) refers to the leg that the assistance is provided more. For example, bilateral additional paretic assistance means that the exoskeleton is providing assistance to both hip joints but provides higher magnitude on the paretic side.

Step length asymmetry was calculated by dividing the paretic side step length by the sum of the paretic and non-paretic side step lengths, where an asymmetry of 0.5 indicates perfect symmetry between the paretic and non-paretic sides. Using five different hip exoskeleton assistance strategies, the participant's Step Length Asymmetry during overground walking was recorded. Assistance types are 1) Unilateral Paretic Assistance, 2) Unilateral Non-Paretic Assistance, 3) Bilateral Equal Assistance, 4) Bilateral Additional Paretic Assistance, and 5) Bilateral Additional Non-Paretic Assistance. The first information (unilateral or bilateral) refers to the leg(s) that the exoskeleton is providing assistance with. For example, unilateral assistance means that the assistance is provided to only one side (zero assistance for the other side). The second information (additional paretic/non-paretic or equal) refers to the leg that the assistance is provided more.

Inclusion Criteria: Age: 18-85 years Had stroke over 6 months prior Greater than 17 on minimental state examination (MMSE) Sit unsupported for a minimum of 30 seconds Follow a 3 step command. Ability to walk without support (a rail as needed is allowed), with a walking speed of at least 0.4 m/s (limited community ambulatory speed) Ability to walk for at least 6 minutes Willingness and ability to participate over a 1-4 hour experiment, with breaks enforced regularly and as needed Ability to transfer (sit-to-stand and stand-to-sit) with no external support (arm rests support allowed) Ability to ambulate over small slopes (3 degrees) and a few steps (6 steps) Exclusion Criteria: Loss of sensation in the legs A complete spinal cord injury History of concussion in the last 6 months History of any severe cardiovascular conditions Severe arthritis Orthopedic problems that limit lower extremity passive range of motion (knee flexion contracture of >10 degrees, knee flexion active ROM 15 degrees) Pre-existing neurological and other disorders such as Parkinson's disease, ALS, MS, dementia History of head trauma Lower extremity amputation Non-healing ulcers of a lower extremity Renal dialysis or end state liver disease Legal blindness or severe visual impairment Uses a pacemaker Has a metal implants in the head region Uses medications that lower seizure thresholds. Lastly, if the subject is participating in another clinical trial and/or subject's condition relating to criteria that, in the opinion of the Principal Investigator (PI), would likely affect the study outcome or confound the results, subject will be excluded from the study.