Optimizing Augmented Reality Cueing Strategies for Freezing of Gait in Parkinson Disease: Interim Results of the ELIMINATE FoG Study
Brendan Baugher1, Ryan Kaya2, Kathryn Scelina2, Debolina Ghosh3, Alexa Amato2, Ken Baker2, Hubert H. Fernandez2, Jay Alberts2, Nathaniel Szewcyzk4, James Liao2
1Ohio University and Cleveland Clinic, 2Cleveland Clinic, 3Case Western Reserve University School of Medicine, 4Ohio University
Objective:

To improve augmented reality (AR) cueing strategies in combatting freezing of gait (FoG) in Parkinson Disease (PD), and to create the first clinically effective non-rhythmic spatial AR cue using a novel, versatile cueing strategy. 

Background:

FoG is a common symptom of PD that can interrupt daily living and cause devastating falls. Previous research demonstrated benefits of AR cueing on FoG, but those cues were not feasible for real-world applications, required attentional demands that cause social disruption, or relied on adherence to a rhythm. Non-rhythmic spatial AR cues are promising in terms of versatility and user-friendliness but have not proven effective in reducing FoG, possibly due to limited AR field-of-view of prior headsets. The Magic Leap 2 (ML2) is a new AR headset with the largest AR field-of-view to date. This hardware presents a fresh opportunity to test spatial AR cueing strategies. PD cueing products have begun to hit market; clinical guidance is needed on their use.

Design/Methods:

36 PD patients will complete walking tasks in AR hallways wearing accelerometers and the ML2. The first intervention will utilize a proven physical cue to test cue-responsiveness. AR cue conditions will be randomized in order: no cue (control), constant, user hand-triggered as-needed, user eye-triggered as-needed, examiner-triggered as-needed. Our cue consists of two concentric circles with radii based on stride length; this shape allows use on turns and straight paths. This is a single-center, within-subjects, unblinded study. FoG and gait parameters will be collected via the ML2, accelerometers, and video, and compared between the test arms.

Results:

Sessions begin November 2022. Data from the first 50% of participants will be presented.

Conclusions:

Initial comparisons of cueing strategies and lessons on ML2 usability (this is the first clinical trial to use this headset) will be used to adapt the study, including a re-analysis of sample size.

10.1212/WNL.0000000000202354