An event-triggered machine learning approach for accelerometer-based fall detection
- Submitting institution
-
Coventry University
- Unit of assessment
- 12 - Engineering
- Output identifier
- 13592091
- Type
- D - Journal article
- DOI
-
10.3390/s18010020
- Title of journal
- Sensors
- Article number
- -
- First page
- 20
- Volume
- 18
- Issue
- 1
- ISSN
- 1424-3210
- Open access status
- Compliant
- Month of publication
- December
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Looking at key roadblocks in widespread uptake and desirability of wearable fall devices, the work challenges current approaches to falls detection; it demonstrates that identification of selective features and consequent real-time computation during fall stages significantly improves detection performance while maintaining battery longevity. Method was validated for all major, publicly available datasets and has extended applicability to IoT contexts where system longevity matters and sensed phenomena are complex. Developed collaboratively with Macquarie University through a co-tutelle. Led to a start-up by the lead author (https://bithouse.id/#home05) that is developing IoT products based on the innovation.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
