Wearable airbag technology and machine learned models to mitigate falls after stroke.
Olivia K Botonis, Yaar Harari, Kyle R Embry and 6 others
PMID 35715823WHAT IT FOUND
A fall detection algorithm trained on stroke survivors' movements detected more falls than one trained on healthy controls.
This suggests commercial airbag devices, which often use general population data, may miss falls in stroke patients because their movement patterns are different.
Key findings
01The model trained on stroke data had significantly higher recall for detecting falls in stroke participants compared to the model trained on control data.
02The advantage of the stroke-trained model was most pronounced for anterior-posterior falls, where it showed significantly higher recall and F1-score compared to the control-trained model.
03Control-trained models performed worse as the number of unstable ambulators in the test group increased, while stroke-trained models remained stable or improved.
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 study was conducted in a laboratory setting with supervised, simulated falls on padded mats, not real-world community falls. The sample size was small (20 stroke participants). The airbag device was modified to prevent deployment, so the actual physical mitigation of impact was not tested. Stroke participants performed fewer falls and non-falls on average than controls due to safety and time constraints. The definition of 'unstable ambulators' was based on gait kinematics, which may not perfectly correlate with actual fall risk.
Declared interests
The study was funded by the Department of Health and Human Services, National Institute on Disability, Independent Living, and Rehabilitation Research. The authors used a commercial device (Wolk Hip Airbag) but modified it for data logging and disabled the deployment mechanism for the study.
The easy way to misread this
Do not assume this study proves the airbag device prevents fractures or injuries. It only tested the algorithm's ability to detect a fall before impact in a lab setting. The actual deployment of the airbag and its protective effect were not evaluated.