Robotic Apparel to Prevent Freezing of Gait in Parkinson Disease

Freezing-of-gait (FoG) in Parkinson Disease (PD) is one of the most vivid and disturbing gait phenomena in neurology. Often described by patients as a feeling of "feet getting glued to the floor," FoG is formally defined as a "brief, episodic absence or marked reduction of forward progression of the feet despite the intention to walk." This debilitating gait phenomena is very common in PD, occurring in up to 80% of individuals with severe PD. When FoG arrests walking, serious consequences can occur such as loss of balance, falls, injurious events, consequent fear of falling, and increased hospitalization. Wearable robots are capable of augmenting spatiotemporal gait mechanics and are emerging as viable solutions for locomotor assistance in various neurological populations. For the proposed study, our goal is to understand how low force mechanical assistance from soft robotic apparel can best mitigate gait decline preceding a freezing episode and subsequent onset of FoG by improving spatial (e.g. stride length) and temporal features (e.g. stride time variability) of walking. We hypothesize that the ongoing gait-preserving effects can essentially minimize the accumulation of motor errors that lead to FoG. Importantly, the autonomous assistance provided by the wearable robot circumvents the need for cognitive or attentional resources, thereby minimizing risks for overloading the cognitive systems -- a known trigger for FoG, thus enhancing the repeatability and robustness of FoG-preventing effects.

Wearable robots are capable of augmenting spatiotemporal gait mechanics and are emerging as viable solutions for locomotor assistance in various neurological populations. Given the breakdown of spatiotemporal gait parameters prior to onset of FoG, we aim to understand how the use of mechanical assistance from a soft robotic apparel can best mitigate gait decline preceding a freezing episode, and subsequent onset of FoG through a multi-day proof-of-concept study. In Aim 1, we will determine the biomechanical mechanisms underpinning the effects of robotic apparel on FoG. We posit that robotic apparel will prevent FoG by supporting natural gait biomechanics and reducing motor errors and gait degradation (i.e., increase stride length, decrease stride variability) known to precede freezing. In Aim 2, we will quantify the impact of robotic apparel in preventing FoG in PD under a variety of walking conditions in a series of controlled laboratory-based experiments. We hypothesize that robotic apparel will be effective in preventing FoG as evidenced by lower percent time spent freezing and lower FoG severity ratio scores (IMU data, video annotation) during walking and turning, resulting in farther walking distances (2-Minute Walk Test) compared to unassisted walking, repeatable across days of testing. Additionally, we hypothesize that robotic apparel will be effective in preventing FoG across various walking contexts (i.e., walking in open spaces, turning, dual-tasking and medication on/off). In Aim 3, we will examine proof-of-concept of robotic apparel to prevent FoG in the home/community during walking, under FoG provoking conditions. We hypothesize that robotic apparel will be effective in preventing FoG, compared to unassisted walking, as evidenced by lower percent time spent freezing and lower FoG severity ratio scores (IMU data, video annotation) during walking in the home/community, including conditions that trigger FoG (e.g., personalized FoG "hotspots). The study will utilize a soft robotic apparel that has previously shown to demonstrate robust, gait-preserving benefits and FoG prevention in a single-subject repeated measures case study. To examine the effectiveness of the intervention using our robotic apparel, this 9-visit study will collect data on amount of time spent freezing, spatiotemporal gait measures, clinical measures, and patient perspectives on the device during different standardized assessments and freeze-provoking activities across multiple environments (i.e. home, lab) and medication states (on, relative off) with and without the robotic apparel assistance.