Assessing the concurrent validity of a gait analysis system integrated into a smart walker in older adults with gait impairments

Christian Werner, Georgia Chalvatzaki, Xanthi S. Papageorgiou, Costas S. Tzafestas, Jürgen M. Bauer, Klaus HauerClinical 2019 · DOI

TL;DR

This study validates a smart walker's integrated gait analysis system against a gold-standard GAITRite system. The smart walker shows good accuracy for temporal gait parameters like stride, swing, and stance time, but less so for spatial parameters like stride length and gait speed, suggesting it's promising for monitoring gait changes but needs refinement for precise absolute measurements.

What Problem Does This Paper Address?

Monitoring gait in older adults with mobility impairments is crucial for rehabilitation and daily living. While wearable sensors exist, they can be uncomfortable and lead to adherence issues. "Smart walkers" offer a potential solution by integrating gait analysis directly into an assistive device, eliminating the need for separate sensors. However, the accuracy and reliability of these integrated systems, especially when compared to established methods like the GAITRite system, need rigorous validation, particularly for the target user group of older adults with gait impairments. This paper aims to fill that gap by assessing the concurrent validity of such a system.

Methods

The researchers compared a novel smart walker's gait analysis system with the well-established GAITRite electronic walkway. Twenty-five older adults (≥65 years) with gait impairments participated. They walked along a 7.8-meter walkway with the smart walker, and their gait was simultaneously recorded by both systems. Key gait parameters measured included stride time, swing time, stance time, stride length, and gait speed. To assess validity, they employed the Bland-Altman method to evaluate bias and limits of agreement, calculated percentage errors (with an acceptable threshold of <30%), and determined intraclass correlation coefficients (ICCs) for both consistency (ICC3,1) and absolute agreement (ICC2,1).

Datasets

Not discussed in the paper. The study used data collected from participants during controlled laboratory experiments.

Key Insights and Ideas

  • Temporal vs. Spatial Accuracy: The smart walker excels at measuring the timing of gait events (stride, swing, stance times) but struggles with the distance covered (stride length, gait speed). This suggests the system is good at detecting changes or consistency in gait rhythm but less reliable for precise absolute distance measurements.
  • Consistency is Key: Even though absolute accuracy for spatial parameters is limited, the smart walker demonstrates good-to-excellent consistency for stride length and gait speed. This implies it can reliably track changes in these parameters over time, which is valuable for monitoring progress in rehabilitation.
  • Reference Point Differences: The discrepancy in spatial measurements is attributed to the different reference points used by each system. The GAITRite system uses heel strikes on the walkway, while the smart walker's laser range finder is positioned higher and measures leg placement, leading to a shorter perceived stride.
  • User-Centric Design: The development of such systems is driven by the need for unobtrusive, continuous gait monitoring in real-world settings, addressing the limitations of wearable sensors.

Key Results

  • Temporal Parameters: Bias for stride, swing, and stance time was minimal (-0.04 to 0.04 seconds), with acceptable percentage errors (8.7%–23.0%). Consistency (ICC3,1) and absolute agreement (ICC2,1) were good to excellent (0.72–0.97).
  • Spatial Parameters: Stride length and gait speed showed higher bias (0.20 m and 0.19 m/s, respectively) and unacceptable percentage errors (31.3%–42.3%). While consistency (ICC3,1) was good to excellent (0.72–0.86), absolute agreement (ICC2,1) was only poor to fair (0.37–0.52).
  • Limits of Agreement: Limits of agreement were considerably narrower for temporal parameters than for spatial ones, reinforcing the difference in accuracy.

Limitations and Future Work

The authors acknowledge several limitations:

  • Controlled Environment: The study was conducted in a controlled laboratory setting, which may not fully represent real-world walking conditions. Future studies should assess validity in less constrained environments.
  • Participant Demographics: The sample was predominantly female, which might limit the generalizability of the results to males. However, the authors did not expect gender to significantly impact concurrent validity.
  • Short Walks: Only short, straight walks were evaluated, limited by the criterion standard.

Future work should focus on:

  • Assessing the system's validity in more naturalistic and less constrained movement situations.
  • Investigating modifications to improve the absolute measurement accuracy of spatial gait parameters.
  • Confirming the system's applicability for long-term gait monitoring in natural environments.

Summary updated

Abstract

The smart walker-integrated gait analysis system has good concurrent validity with the GAITRite system for measuring temporal but not spatial-related gait parameters in potential end-users of the smart walker. Stride length and gait speed can be measured with good consistency, but with only limited absolute accuracy.
BibTeX
@inproceedings{werner2019assessing,
  author = {Christian Werner and Georgia Chalvatzaki and Xanthi S. Papageorgiou and Costas S. Tzafestas and Jürgen M. Bauer and Klaus Hauer},
  title = {{Assessing the concurrent validity of a gait analysis system integrated into a smart walker in older adults with gait impairments}},
  booktitle = {Clinical},
  year = {2019},
  doi = {10.1177/0269215519852143},
}
Figure