Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions.
Sina Mehdizadeh, Hoda Nabavi, Andrea Sabo and 3 others
PMID 34526074WHAT IT FOUND
Video pose tracking accurately measures step time and cadence in older adults, but fails to reliably measure step width or variability.
Clinicians should use these tools only for temporal gait metrics, not for assessing lateral stability or step consistency.
Key findings
01Temporal gait measures (cadence and step time) showed high to very high correlations with motion capture, while spatial measures (step width and margin of stability) showed low or moderate correlations.
02Averaging gait variables over all steps improved correlations, but spatial measures remained non-significant, suggesting the sample size was insufficient for these specific metrics.
03There were minimal differences in validity between the three tested pose tracking algorithms (AlphaPose, OpenPose, Detectron) or between camera heights and viewing angles.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The sample size was very small (n=11), which reduced statistical power, particularly for spatial and variability measures. Participants were all able to walk independently for 20m and had no recent falls, so results may not apply to those with significant mobility impairments. The study only analyzed straight-line walking; turns and stops were excluded, limiting ecological validity. The pose tracking algorithms were not specifically designed for quantitative gait analysis, which may explain the poor spatial accuracy.
Declared interests
Funding was provided by Mitacs, AGE-WELL, Riverview Health Center, Walter and Maria Schroeder Institute for Brain Innovation and Recovery, and Institute of Aging.
The easy way to misread this
Do not assume that because video tracking accurately captures cadence, it can also accurately measure step width or stability. The study found spatial measures were poorly correlated with gold-standard motion capture and often non-significant, meaning these specific metrics cannot be trusted for clinical decision-making.