Augmenting Balance and Spatial Awareness in Parkinson’s Disease: A Wearable Assistive Tail

More than 10 million people around the world are estimated to be living with Parkinson’s Disease (PD), a brain disorder that progressively damages dopamine-producing neurons and increases in risk with age, framed by symptoms such as tremor, stiffness and slowed movement (Figure 1) [1]. In Canada alone, an average of 38 people are diagnosed with Parkinsonism daily, making it one of the most prevalent diseases in the world [2]. Studies show that individuals with PD have 1.5-2X higher fatality rate not due to the disease itself, but often because falls and injuries are fatal [3, 4].

Currently, treatments regarding PD aim to reduce symptoms and decrease the risk of falls. Although traditional methods such as compression socks may alleviate certain symptoms, they do not provide active gait correction and may be ineffective at augmenting balance. Several novel studies have shown that robotic tails provide significantly better support for movement in patients with PD, but these devices remain largely experimental [5, 6]. Furthermore, these devices are powered by advanced pneumatic and hydraulic systems, making materials very expensive and often inaccessible to the larger population.

Thus, we had 5 primary design objectives for a wearable device that reduces fall risk:

An affordable, modular system that applies basic physical and biological principles

Replicable and easily customizable in accordance with patient needs to ensure comfort

Accurately model the individual using center of mass calculations

Live biometric tracking and in-app alerts to ensure patient safety

Integrate accessible, controllable mechanisms to achieve gait correction

Preliminary Testing

A preliminary experiment was carried out to identify problematic symptoms in PD patients, and several abnormalities were noticed:

Externally rotated hips and feet, producing a duck-footed stance. This is in accordance with PD, as patients have weaker hip flexor and rotator muscles [7].

Shuffling and flat-footed gait, producing an unstable base of support.

Hunched posture gives patients a false sense of security, but increases fall risk.

All symptoms detected were in accordance with clinical trials of PD in the larger population.

Tested masses from 200g to 2000g to find a suitable counterbalance range.

Conducted three trials per weight on each patient.

Attached 3D-printed designs using a tubing system.

Had patients walk 10m during testing.

Collected patient comfort feedback.

Measured sway with and without added weight to assess impact.

Findings:

200-400g: patients felt no effect when attached to their backs.

400g-800g: optimal weight for comfort, but did not achieve the full correction effect.

800g-1200g: saw effective correction while still maintaining patient comfort.

1200g-2000g: patients found this to be uncomfortable despite it having greater corrective power.

Note: We noticed that patients who were smaller in stature preferred less weight.

Patient Demographics (Table 1)

We found four suitable patients aged 54-91 with the following demographics and stages of PD

Design (Fig. 2, 3)

Hardware Components

An ultrasonic sensor was mounted on harness.

Haptic feedback motors were embedded on the harness at the left & right lower back positions.

The accelerometer is integrated into the harness.

Data collected by the above sensors transmitted to the microcontroller through wires.

Software Components (Fig. 4)

Code will be implemented to deliver adaptive postural stabilization and record quantitative stability data. This will be split into several components:

Initialization of the device

Real-time sway detection

Perturbation classification

Tail actuation and haptic feedback response

Postural stability logging

Universally accepted clinical measurements for PD postural and balance control were primarily selected to measure tail efficacy [8, 9, 10]. All sensor values are also implemented in-app for medical professionals, serving as logged data for examination in case of emergencies during and following potential falls.

Safety Precautions

Allowed ample patient rest between trials

Did not cause patients and emotional or physical distress

Ensured viable support around patient in case of emergency

Sway Range Measurement (Fig. 5)

The ultrasonic sensor captures distance readings at 40 kHz throughout each trial.

Each reading was compared against the baseline distance to compute an instantaneous sway value

Positive values indicated forward displacement, and negative values indicated backward displacement

Sway range was calculated as the difference between the max and min sway values recorded over the trial window

Perturbation Recovery Time (Fig. 5)

A perturbation event is defined as any sway deviation exceeding 3.0 cm from baseline based on our preliminary testing.

Perturbation recovery time is calculated as the elapsed time between perturbation onset and the moment sway returns within the threshold boundary

Upon detection, the microcontroller simultaneously triggers the tail actuation response described and the haptic feedback response, and begins logging the recovery interval

Perturbation Detection Speed (Fig. 5)

When a perturbation is detected by the ultrasonic sensor, the microcontroller begins logging the three-axis acceleration data from the accelerometer at 100 Hz.

Perturbation detection speed is defined as the elapsed time between perturbation onset and the moment the resultant acceleration magnitude reaches its peak value

Shorter detection speed in the without-tail condition indicated a more abrupt, unmitigated perturbation response

a longer detection speed in the with-tail condition indicates that the tail's mechanical counterbalancing is reducing the event’s abruptness and giving the participant more time to adjust

From these metrics shown in Figure 6, we found:

The optimal counterweight configuration is strongly participant-dependent rather than universally scalable

P1 benefited most from the heaviest configuration across all metrics, consistent with strength to manage the 1200g load

P2 and P3 both performed best at 800g across the majority of metrics, heavier configurations producing worse results

Particularly notable in P2's non-linear gait response and P3's consistent over-correction pattern

P4's overall best configuration was 400g, with both 800g and 1200g producing deterioration, likely due to low body mass and PD stage

Thus, clinical prescription of the wearable assistive tail should involve individualised weight calibration based on participant age, body mass, and mobility profile rather than a standardised configuration.

Conclusions and Results

Postural stability metrics improved across all areas:

Mean sway range reduced by 40.4% (4.50 cm → 2.68 cm) at respective best configurations

Mean perturbation recovery time improved by 33.8% (2,099 ms → 1,390 ms) at 800g

Mean perturbation detection speed increased by 132% (36.8 ms → 85.5 ms), indicating that the tail helped increase the rate of stability correction

User experiences were largely favourable throughout:

Participants reported the device as comfortable to wear during all three tasks and found visible improvements

Specifically, P2 found that they were able to balance longer (5-10s more on each foot)

P4 found that they were able to walk several (>5-10) meters (being unable to walk at all or only <2m without aid)

Stabilization effects felt natural, so users did not need to consciously engage with the device

Haptic feedback was intuitive, assisted by the easy user interface of the mobile app

Broader Significance

Wearable counterweight-based postural stabilization for Parkinson's disease is a largely unexplored intervention space, as existing solutions focus primarily on medication and physiotherapy, where this project focuses on the modular components and ensuring that the tail stays affordable.

Significant commercial potential if refined and scaled, as the tail mechanism could be integrated into existing mobility aids or adapted for other balance-impairment conditions such as stroke recovery or vestibular disorders. Furthermore, users can easily customize their tail with removable weight segments and belt attachments, with all parts able to be batch- or mass-produced.

In future development iterations of the tail, further scientific improvements will be highly relevant:

Larger, more diverse participant groups enable the analysis of data correlations involving metrics like disease progression

Long-term adaptation and mobility tests demonstrate its effectiveness in varying environments beyond a controlled walking space

Usage in conjunction with existing mobility aids, such as walking canes, enables new metrics such as step length and walking speed

The highly personalizable, modular design enables various future additions that could improve its holistic spatial awareness:

GPS and 911 auto-dialling for severe falls and location tracking

Integration with voice prompts and audio cues

Firstly, we would like to acknowledge the Team Calgary delegates for their guidance throughout this project, giving us valuable feedback for improvements.

We also want to acknowledge Ms. Trainor for her guidance as our school supervisor for CYSF and for her support of this project.

Lastly, we would like to thank Lu's Clinic for helping us in sourcing suitable volunteers for our experiment. This project could not have been possible without the support of these volunteers and the clinical setting.

We also want to acknowledge our parents for their moral support throughout the project and all the authors of the works we have referenced and learned from.

[1] What is Parkinson’s? | Parkinson’s Foundation. (n.d.). Retrieved 30 April 2026, from https://www.parkinson.org/understanding-parkinsons/what-is-parkinsons

[2] Canada, P. H. A. of. (2022, April 26). Data blog on parkinsonism in canada, including parkinson disease—Canada. Ca[Datasets;statistics;education and awareness]. https://health-infobase.canada.ca/datalab/parkinson-blog.html

[3] Ryu, D. W., Han, K., & Cho, A. H. (2023). Mortality and causes of death in patients with Parkinson's disease: a nationwide population-based cohort study. Frontiers in neurology, 14, 1236296. https://doi.org/10.3389/fneur.2023.1236296

[4] Ph.D, R. G., MD. (2022, February 8). End-stage parkinson’s disease & death | apda. American Parkinson Disease Association. https://www.apdaparkinson.org/article/death-parkinsons-disease-3/

[5] Wearable robot helps man with Parkinson’s disease to walk. | Harvard Biodesign Lab. (n.d.). Retrieved from https://biodesign.seas.harvard.edu/publications/wearable-robot-helps-man-parkinson%E2%80%99s-disease-walk

[6] 'Robot tail' could help reduce risk of falling for elderly, say Japanese scientists | The Telegraph. (2019 August 5). Retrieved from https://www.telegraph.co.uk/technology/2019/08/05/robot-tail-could-help-reduce-risk-falling-elderly-say-japanese/

[7] Mak, M. K., Wong-Yu, I. S., Shen, X., & Chung, C. L. (2017). Long-term effects of exercise and physical therapy in people with Parkinson disease. Nature Reviews Neurology, 13(11), 689–703. https://doi.org/10.1038/nrneurol.2017.128

[8] Johansson, J., Nordström, A., Gustafson, Y., Westling, G., & Nordström, P. (2017). Increased postural sway during quiet stance as a risk factor for prospective falls in community-dwelling elderly individuals. Age and Ageing, 46(6), 964–970. https://doi.org/10.1093/ageing/afx083

[9] Gerards, M. H. G., McCrum, C., Mansfield, A., & Meijer, K. (2017). Perturbation‐based balance training for falls reduction among older adults: Current evidence and implications for clinical practice. Geriatrics & Gerontology International, 17(12), 2294–2303. https://doi.org/10.1111/ggi.13082

[10] Kwon, K. Y., You, J., Kim, R. O., Lee, E. J., Lee, J., Kim, I., Kim, J., & Koh, S. B. (2024). Association Between Baseline Gait Parameters and Future Fall Risk in Patients With De Novo Parkinson's Disease: Forward Versus Backward Gait. Journal of clinical neurology (Seoul, Korea), 20(2), 201–207. https://doi.org/10.3988/jcn.2022.0299

Additional Images

National initiative for care of the elderly. NICE. (n.d.). https://www.nicenet.ca/