Tug-vimu: a GoPro-based mobility assessment bridging the gap between technology and clinical practice.
Sara Houde, Camille Martin, Karina Lebel
PMID 41310629WHAT IT FOUND
In 16 healthy adults, a GoPro chest-camera TUG system detected 98% of gait steps and measured step length and duration reliably at slow and regular speeds.
Some turning and sit-to-stand metrics missed predefined error targets.
Key findings
01The GoPro system segmented the TUG with a mean test-duration error of 0.36 seconds, and phase errors of 0.37 seconds for sit-to-stand and 0.15 seconds for turning.
02Gait step detection was 98% reliable, and step length and step duration showed excellent relative reliability at slow and regular speeds, but step duration reliability was moderate at fast speed.
03Not all mobility metrics met accuracy targets: the sit-to-stand lean angle limits of agreement ranged from -17 degrees to 8 degrees, and maximum turn velocity failed the lower confidence interval threshold.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
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
Only 16 healthy adults were tested, including 8 participants over 60 years, so the results do not show accuracy in patients with mobility impairments. Thirty-five of 288 trials were excluded for poor execution or algorithm errors, leaving 253 trials. Some parameters did not meet predefined accuracy targets, including lean angle confidence interval, maximum turn velocity, toe-off at fast speed, and step length at regular speed. ArUco marker tracking was lost for an average of 19.3% of each test, mainly during sit-to-stand and turning. The study used a cone for turning, which may change turning strategy compared with floor tape. Fast walking was less representative of the population that would use a mobility test clinically.
Declared interests
The study was funded by the Natural Sciences and Engineering Research Council of Canada and Canada Research Chairs. The supplied text does not state a competing-interest declaration.
The easy way to misread this
Do not conclude that this GoPro system is ready for clinical mobility or fall-risk assessment. It was tested only in 16 healthy adults, and several key parameters did not meet accuracy targets.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →