Individuals with Parkinson’s disease who experience freezing of gait (FoG) — the sudden inability to continue walking — had fewer and shorter FoG episodes when receiving spatial cues through augmented reality (AR) headsets.
Advertisement
Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services. Policy
With AR, digital objects are placed in the user’s environment and these digital objects can be interacted with, or in this case, provide spatial cues. Results of the ELIMINATE FoG trial, published recently in Clinical Parkinsonism & Related Disorders, also reveal that participants have different preferences for how AR cues were delivered.
The study offers new ways AR may be used therapeutically and, perhaps eventually in daily life, to relieve a drug-resistant symptom of PD that can dramatically diminish quality of life and safety.
Previous studies have shown people exhibited better walking patterns with the help of AR cueing but they did not impact or reduce freezing, says Cleveland Clinic neurologist and project leader, author James Liao, MD, PhD.
“This is the first study that demonstrates the feasibility of using AR cues to improve FoG,” he says. “Not surprising, patients liked the technology and have preferred types of stimuli to affect FOG. This patient-specific information is shaping how we are creating the next generation of the software for routine clinical use.”
Challenges of FoG
Postural instability and gait dysfunction reduce quality of life for people with Parkinson’s, says Jay Alberts, PhD, Vice Chair of Innovation in the Neurological Institute. During freezing, “the person feels like their feet are glued to the floor,” says Dr. Alberts. “This is a problem for two reasons: First, and most obvious, the person is frozen, and second, because if you’re walking and suddenly if your feet feel glued to the floor and you stop, your momentum continues; thus freezing is a precursor to falling.
“If a person is afraid of freezing, they avoid community settings,” Dr. Alberts adds. “This avoidance results in isolation. We are hopeful this technology will be used to reduce FOG and give patients confidence to remain in or return to community environments.”
Introducing external cues, such as laying tape stripes along a path on the floor, have long been used to try to interrupt the neurological dysfunctions suspected of causing FoG. A limited number of investigations into AR cueing have led to inconclusive results.
For ELIMINATE FoG, Drs. Alberts and Liao collaborated to design the AR cueing strategies. Their hypothesis was that cues that required more cognitive effort to activate would be less effective.
Methods and results
For the study, 36 people with Parkinson’s-related FoG tested six cueing conditions while they walked a holographic hallway wearing an AR headset.
The tested conditions were:
No cue. Participants walked the course with no cues, and tended to get stuck in tight spaces such as turns.
Physical cue. Parallel lines of white tape were affixed to the ground along the course.
Constant AR cueing. Two concentric rings appeared on the floor at all times while participants walked the course.
User-controlled AR cueing. Concentric rings appeared when the participant clicked a button. The rings stayed where they were the instant the button was pressed, and participants walked past them, at which point the rings disappeared.
Observer-controlled AR cueing. The observer would click the button to make the rings appear. This was intended to mimic a yet-to-be-created algorithm that could anticipate when a cue was needed.
Eye-controlled AR cueing. The participant looked down at where the ring cues were needed, and they appeared.
The researchers introduced three levels of difficulty. The first featured a doorway and two column-like obstacles. The second included a cognitive task directed through the headset, such as completing a simple math or spelling problem. The third included the cognitive task, plus a second doorway and two more obstacles.
Kinematic and inertial measurement unit data were collected by the AR headset and wearable sensors, respectively. Gait progress was recorded by video, and patients participated in a post-test survey about their experiences and their cueing preferences.
Primary outcomes were percentage of time frozen and freeze rate — the number of freezes occurring each minute. Secondary outcomes included gait metrics and participant survey responses.
Study results
All 36 participants completed at least one minute of testing for each condition. A total of 1,285 episodes of FoG were observed for all participants. Although more than 30% of the walking course was straight and contained no obstructions, only 5.6 % of FoG episodes took place there.
Among all freezing episodes:
- 98.08 % had cue activations in the observer-controlled condition
- 75.34 % had activations in the eye-controlled condition
- 63.86 % in the hand-controlled conditions
The no-cue, physical cue and constant cue conditions did not require activation. “Across the board, we found that the constant-on cue reduced how often freezes occur,” says Dr. Liao. “The intermittent cues tend to reduce how long the freeze lasts, but they don’t reduce the number of freezes.”
The observer-controlled cueing reduced the percentage of time a participant was frozen, and the constant cue decreased FoG incidence compared to controls.
For the 28 participants who identified an AR cue as their preferred mode, the specific condition they preferred significantly reduced the percentage of time frozen and freeze rate compared to controls.
Next steps
The study findings were surprising, says Dr. Liao, and also suggest paths for future research.
“We expected that the low cognitive burden cues would work the best, so that would be observer-controlled or perhaps the physical cue. But the results didn’t fall along those lines,” he says. “The reduction of the freeze rate with the constant cue, and the reduction in freeze duration with the intermittent cues, were quite interesting.”
A potential next step is to build an algorithm capable of replacing the observer-controlled queuing. More research also is needed to determine how AR cueing works in a home setting, whether effectiveness can be sustained over time, and if regular use might trigger durable neuroplastic changes.
These questions will become even more relevant as technology companies bring lighter, more flexible “smart” eyeglasses to the marketplace.
“As more and more consumer-grade augmented reality devices are made, we will have an opportunity to integrate AR cueing modes into real life,” says Dr. Alberts. “So we could think about providing something like that to an individual with Parkinson’s, and it could have an accelerometer and a gyroscope to identify where they are and to predict, when a freeze may occur. The AR glasses would deliver a cue to prevent that freezing episode.”
This technology has been licensed to Strolll Limited for clinical distribution.