Auto detection and segmentation of daily living activities during a Timed Up and Go task in people with Parkinson's disease using multiple inertial sensors.
Hung Nguyen, Karina Lebel, Patrick Boissy and 3 others
PMID 28388939WHAT IT FOUND
After adjustments for Parkinson's movement patterns, wearable sensor algorithms detected standing, sitting, walking and turning correctly in all 432 tested activity instances during a Timed Up and Go task in 12 people with early disease.
Key findings
01Using the original algorithms developed in healthy older adults, walking was detected with 91.6% sensitivity in the 10 m Timed Up and Go task and 88.9% sensitivity in the 5 m task.
02After modifying the sensors and algorithms for Parkinson's movement patterns, all activities were detected with 100% sensitivity and specificity in both tasks.
03After re-optimizing filters, automated transition timing differed from examiner timing by an average of 453 ms in the 5 m task and 345 ms in the 10 m task.
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 12 participants were tested, all with early Parkinson's disease and independent daily living, so the results may not apply to more advanced disease or people who cannot perform a Timed Up and Go task. The task was supervised and scripted, so the study does not show that the algorithms work in free-living home or community environments. The algorithms were modified using 10 m trial data and then applied to 5 m trials; the paper says this does not address transferability to all people with Parkinson's disease regardless of stage or type. The paper notes the 17-sensor suit is not the most economical system, although only 4 sensors were needed after optimization; cost and usability were not evaluated. The study measured detection and timing accuracy, not clinical outcomes, mobility improvement or treatment decisions.
Declared interests
The study was funded by the Canadian Institutes of Health Research.
The easy way to misread this
Do not conclude that wearable sensors improve Parkinson's disease mobility or patient care. The 100% detection accuracy was for identifying task segments in 12 people with early Parkinson's disease during a supervised Timed Up and Go task, not for treatment effect or free-living use.