Periodic Reporting for period 1 - BalancingACT (Balancing Adaptive Cooperative Technology)
Période du rapport: 2023-08-01 au 2025-07-31
Balance and gait impairments are among the most disabling consequences of neurological and age-related conditions. They affect millions of people in Europe and worldwide, limiting independence, reducing participation in society, and increasing the risk of falls and fall-related injuries. Falls are a leading cause of hospitalization, long-term care admission, and loss of autonomy, making balance disorders a critical public health and societal challenge.
Current solutions, such as canes or walkers, provide partial support, although there is room for improvement. In addition to requiring use of the hands, these devices can be bulky, stigmatizing, or insufficient in dynamic and real-life environments. Rehabilitation can improve motor control, but the effectiveness is limited by the nature of the impairments: some conditions are degenerative, leading to progressive decline, and the specific pattern of impairments varies greatly between individuals. There is, therefore, a pressing need for innovative, user-friendly technologies that can actively support balance, enhance rehabilitation, and ultimately reduce fall risk.
Within this context, the BalancingACT project set out to explore how lightweight, wearable robotic devices (i.e. gyroscopic actuators capable of modulating whole-body stability) can be harnessed to improve balance and gait. The primary goal of this project was to integrate insights from motor learning and neuromechanics research in order to effectively adapt both the wearable robotic devices as well as the human using the device.
Current solutions, such as canes or walkers, provide partial support, although there is room for improvement. In addition to requiring use of the hands, these devices can be bulky, stigmatizing, or insufficient in dynamic and real-life environments. Rehabilitation can improve motor control, but the effectiveness is limited by the nature of the impairments: some conditions are degenerative, leading to progressive decline, and the specific pattern of impairments varies greatly between individuals. There is, therefore, a pressing need for innovative, user-friendly technologies that can actively support balance, enhance rehabilitation, and ultimately reduce fall risk.
Within this context, the BalancingACT project set out to explore how lightweight, wearable robotic devices (i.e. gyroscopic actuators capable of modulating whole-body stability) can be harnessed to improve balance and gait. The primary goal of this project was to integrate insights from motor learning and neuromechanics research in order to effectively adapt both the wearable robotic devices as well as the human using the device.
The objectives of this project were to address different aspects of human-robot interaction that are critical to enhancing human-machine performance. The first objective was to identify specific metrics that meaningfully describe balance and stability and can also be modulated with our device. The second work package addressed the human side of the system, leveraging motor learning strategies (often well-researched in upper limb movements) for balance and walking tasks. The third objective combined these findings to determine optimal levels of assistance.
Two major results on this topic were achieved as part of a collaboration with another funded project. In the first study, healthy participants wore the device, in different assistance levels, for different tasks that challenged standing and walking balance. These findings not only illustrated the positive effect such a device could have on balance, but also provided justification for further study into other aspects of human-robot interaction. For example, training and motor learning, participant preference, and perception of assistance (i.e. if they could distinguish between assistance and a placebo/sham condition) all contributed to our final results. My contributions on this project include writing and revising the final manuscript (particularly on the motor learning and protocol recommendations), and the findings have led to two ongoing projects funded by the BalancingACT grant: one seeking to characterize the neuromechanical response to mechanical perturbations from the device, decoupling the responses elicited during perceived perturbations (i.e. sham perturbations); and the second building on the motor learning findings to determine how well the backpack performs as an assistive aid compared to more conventional forms of assistance (i.e. force fields).
The second collaborative study built on these findings to assist balance and gait in people with degenerative ataxia, across different functional tests, demonstrating the real-world applicability for this powerful technology. My role in this study was focused on protocol development and aiding in the manuscript preparation.
During the course of the BalancingACT project, we have also completed studies more specifically aimed at motor learning in human-robot interactions. I have written a book chapter on variability in the context of motor learning in gait and co-organized a conference workshop to highlight the research being done in this area.
Additionally, I have contributed to a long-term data collection to quantify motor development in terms of gait stability and balance. Together with a team of therapists, students, and rehabilitation medicine researchers, we have created and refined a set of tasks and metrics that can track motor development, both for typically developing children as well as children with neuromotor disorders.
Two major results on this topic were achieved as part of a collaboration with another funded project. In the first study, healthy participants wore the device, in different assistance levels, for different tasks that challenged standing and walking balance. These findings not only illustrated the positive effect such a device could have on balance, but also provided justification for further study into other aspects of human-robot interaction. For example, training and motor learning, participant preference, and perception of assistance (i.e. if they could distinguish between assistance and a placebo/sham condition) all contributed to our final results. My contributions on this project include writing and revising the final manuscript (particularly on the motor learning and protocol recommendations), and the findings have led to two ongoing projects funded by the BalancingACT grant: one seeking to characterize the neuromechanical response to mechanical perturbations from the device, decoupling the responses elicited during perceived perturbations (i.e. sham perturbations); and the second building on the motor learning findings to determine how well the backpack performs as an assistive aid compared to more conventional forms of assistance (i.e. force fields).
The second collaborative study built on these findings to assist balance and gait in people with degenerative ataxia, across different functional tests, demonstrating the real-world applicability for this powerful technology. My role in this study was focused on protocol development and aiding in the manuscript preparation.
During the course of the BalancingACT project, we have also completed studies more specifically aimed at motor learning in human-robot interactions. I have written a book chapter on variability in the context of motor learning in gait and co-organized a conference workshop to highlight the research being done in this area.
Additionally, I have contributed to a long-term data collection to quantify motor development in terms of gait stability and balance. Together with a team of therapists, students, and rehabilitation medicine researchers, we have created and refined a set of tasks and metrics that can track motor development, both for typically developing children as well as children with neuromotor disorders.
A similar device was previously developed by my supervisor that showed promise in assisting balance for healthy adults and people with chronic stroke. The newer version of the device is smaller and sleeker, improving usability and increasing the likelihood of use outside of the lab. Because the gyroscopic actuators in the current version are smaller than the previous iteration, two actuators were used, not only doubling the capacity of the device, but also expanding its utility as a research tool. For example, by having two actuators, we can design a placebo/sham mode, which is an important aspect of rehabilitation research that is otherwise difficult to implement in robotic aids. The hardware was developed in collaboration with researchers and engineers funded by other projects, as the goal of the BalancingACT was to identify strategies and facilitate human-robot co-adaptation. Towards this end, we have emphasized the utility of the placebo/sham condition and incorporating motor learning principles into our experimental designs in these and ongoing experiments.
The gyroscopic actuators are already being deployed in novel research applications, including walking robots, and there is significant interest from the robotics community to collaborate. I will be starting my own lab in the coming year, and I will use the device as a research tool, leveraging current collaborations as well as building new research relationships. As a research tool, the possible applications are vast and currently underexplored, although we have laid out recommendations for protocol adjustments in related publications.
The gyroscopic actuators are already being deployed in novel research applications, including walking robots, and there is significant interest from the robotics community to collaborate. I will be starting my own lab in the coming year, and I will use the device as a research tool, leveraging current collaborations as well as building new research relationships. As a research tool, the possible applications are vast and currently underexplored, although we have laid out recommendations for protocol adjustments in related publications.