Innovative Biofeedback Interface for Enhancing Stroke Gait Rehabilitation

This study will conduct a preliminary evaluation of and obtain user data on a novel game-based visual interface for stroke gait training. Study participants will complete gait biomechanics evaluations comprising exposure to gait biofeedback systems in an order determined by block randomization (3 blocks). Participants will be exposed to 3 types of biofeedback interfaces, as well as control condition: the anterior ground reaction forces (AGRF) biofeedback game (projector-screen display, non-virtual reality VR]) VR version of the biofeedback game (head-mounted AR display) traditional, non-game-based biofeedback interface

Stroke is the leading cause of adult disability. Even after discharge from rehabilitation, residual gait deficits are prevalent in stroke survivors, leading to decreased walking speed and endurance. Because gait dysfunctions limit community mobility, stroke survivors and rehabilitation clinicians consider restoration of walking a major goal of rehabilitation. Several challenges and research gaps limit the effectiveness of current clinical gait rehabilitation practices. While there is consensus that stroke survivors benefit from gait rehabilitation, agreement is lacking on which specific training interventions are most efficacious. The long-term goal of this proposal is to address these challenges by developing personalized, engaging, salient gait training treatments founded on evidence from neuroscience, biomechanics, motor learning, and gaming. Real-time biofeedback is a promising gait training intervention for targeting specific biomechanical impairments. Biofeedback can enhance an individual's awareness of the impairment targeted during gait training, enabling self-correction of aberrant gait patterns. In response to treadmill training combined with visual and auditory real-time biofeedback, able-bodied individuals can increase AGRF unilaterally for the targeted limb. Thus, AGRF biofeedback may be a beneficial strategy to target unilateral propulsive deficits in people post-stroke. Incorporation of gaming interfaces for gait biofeedback can increase patient motivation, distract participants from fatigue or boredom, and encourage greater numbers of repetitions during gait training. The long-term goal of this study is to develop a more engaging, motivating gait biofeedback methodologies specifically designed for post-stroke gait training. The researchers aim to address a major challenge for rehabilitation clinicians - to make gait training appealing and meaningful so that patients engage in sufficient repetitions, intensity, and challenge to maximize therapeutic effectiveness. The study premise is that post-stroke individuals will demonstrate greater engagement, motivation, and therapeutic benefits during gait training sessions involving biofeedback when training incorporates intuitive, entertaining, game-based interfaces. Outcomes of the study will include measures of participant engagement, user-reports and survey-responses on motivation, fatigue, game characteristics, and adverse effects (e.g. nausea, dizziness) during game exposure. In addition to this clinical trial with stroke survivor participants, data about the game interface will be collected by having able-bodied neuro-rehabilitation clinicians try the 3 types of biofeedback interventions.

12 post-stroke and 10 able-bodied individuals will complete gait trials with 3 different biofeedback interfaces.

The study outcomes will comprise objective biomechanical measures derived using motion capture, which will be processed by a lab personnel

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: game without VR, game with VR, traditional interface

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: game without VR, traditional interface, game with VR

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: game with VR, game without VR, traditional interface

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: game with VR, traditional interface, game without VR

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: traditional interface, game without VR, game with VR

Post-stroke participants randomized to receive the three gait training biofeedback interfaces in the order of: traditional interface, game with VR, game without VR

During the AGRF biofeedback game without virtual reality (VR) participants will walk on a split-belt treadmill with a projector or screen display. Participants will be receiving real-time information or feedback about ongoing gait performance parameters while they are walking. The participants will be exposed to the AGRF biofeedback game without VR for 4 minutes.

During the AGRF biofeedback game with virtual reality (VR) participants will walk on a split-belt treadmill with a head-mounted VR display. Participants will be receiving real-time information or feedback about ongoing gait performance parameters while they are walking. The participants will be exposed to the AGRF biofeedback game with VR for 4 minutes.

For the traditional non-game biofeedback, the visual display comprises a horizontal line graph with a moveable cursor that represents the current measured value of antero-posterior ground reaction force for the targeted leg. The auditory feedback comprises an audible "beep" produced every time the cursor entered the target range. Participants will be exposed to the traditional biofeedback interface for 4 minutes.

Boredom, motivation, etc will be assessed by 8-point visual analog scale asking participants to rate how bored they felt during the walk, where 1 is not bored and 8 is very bored.

Paretic leg or targeted leg peak pushoff force (peak AGRF) will be calculate from GRF data. The success rate of the targeted training task will be assessed as the percentage of steps for which the target paretic AGRF was achieved.

The NASA task load index (NASA - TLX) will be used to evaluate user reports of perceived effort in different domains

The HRERS will be used to assess participant engagement during each biofeedback bout. The HRERS is a 5-item instrument where clinicians indicate how engaged the participant was during a rehabilitation session on a 6- point scale ranging from 1 (never) to 6 (always). Total scores range from 5 to 30 with higher scores indicating increased engagement.

Inclusion Criteria for Post-Stroke Participants: chronic stroke (>6 months post stroke) ambulatory with or without the use of a cane or walker able to walk for 2 minutes at the self-selected speed without an orthoses resting heart rate 40-100 beats per minute Exclusion Criteria for Post-Stroke Participants: cerebellar signs (ataxic ("drunken") gait or decreased coordination during rapid alternating hand or foot movements history of lower extremity joint replacement inability to communicate with investigators neglect/hemianopia, or unexplained dizziness in last 6 months neurologic conditions other than stroke orthopedic problems in the lower limbs or spine (or other medical conditions) that limit walking or cause pain during walking Criteria for able-bodied participants and neuro-rehabilitation clinicians: -No medical or musculoskeletal conditions limiting walking

Individual participant data that underlie results, after de-identification, will be available for sharing.

Individual participant data will be available for sharing beginning 9 months and ending 36 months after the article publication.

Data will be available for sharing with investigators whose proposed used of the data has been approved by an independent review committee identified for this purpose. Data will be shared for the purposes of achieving the aims in the approved proposal. Proposals should be directed to trisha.m.kesar@emory.edu. To access the data, requestors will need to sign a data access agreement.