RT info:eu-repo/semantics/article
T1 A simulator to support machine learning-based wearable fall detection systems
A1 Collado Villaverde, Armando
A1 Cobos Maestre, Mario
A1 Muñoz Martínez, Pablo
A1 Fernández Barrero, David
K1 Fall detection
K1 Machine learning
K1 Simulation
K1 Wearable devices
K1 Informática
K1 Computer science
AB People’s life expectancy is increasing, resulting in a growing elderly population.That population is subject to dependency issues, falls being a problematic one due to the associatedhealth complications. Some projects are trying to enhance the independence of elderly people bymonitoring their status, typically by means of wearable devices. These devices often feature MachineLearning (ML) algorithms for fall detection using accelerometers. However, the software deployedoften lacks reliable data for the models’ training. To overcome such an issue, we have developed apublicly available fall simulator capable of recreating accelerometer fall samples of two of the mostcommon types of falls: syncope and forward. Those simulated samples are like real falls recordedusing real accelerometers in order to use them later as input for ML applications. To validate ourapproach, we have used different classifiers over both simulated falls and data from two publicdatasets based on real data. Our tests show that the fall simulator achieves a high accuracy forgenerating accelerometer data from a fall, allowing to create larger datasets for training fall detectionsoftware in wearable devices.
PB MDPI
SN 2079-9292
YR 2020
FD 2020-11-03
LK http://hdl.handle.net/10017/48769
UL http://hdl.handle.net/10017/48769
LA eng
NO Junta de Comunidades de Castilla-La Mancha
DS MINDS@UW
RD 19-abr-2024