Triaxial Accelerometer Located on the Wrist for Elderly People’s Fall Detection
The loss of motor function in the elderly makes this population group prone to accidental falls. Actually, falls are one of the most notable concerns in elder care. Not surprisingly, there are several technical solutions to detect falls, however, none of them has achieved great acceptance. The popularization of smartwatches provides a promising tool to address this problem. In this work, we present a solution that applies machine learning techniques to process the output of a smartwatch accelerometer, being able to detect a fall event with high accuracy. To this end, we simulated the two most common types of falls in elders, gathering acceleration data from the wrist, then applied that data to train two classifiers. The results show high accuracy and robust classifiers able to detect falls.
KeywordsFall detection Accelerometer Machine learning Classification Supervised learning Care for the elderly
The authors thank the contribution of Isabel Pascual Benito, Francisco López Martínez and Helena Hernández Martínez, from Department of Nursing and Physiotherapy of the University of Alcalá, for their help designing and supervising the simulated falls procedure. This work is supported by UAH (2015/00297/001), JCLM (PEII-2014-015-A) and MINCECO (TIN2014-56494-C4-4-P).
- 2.Noury, N., Fleury, A., Rumeau, P., Bourke, A.K., Laighin, G.O., Rialle, V., Lundy, J.E.: Fall detection - principles and Methods. In: 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 1663–1666 (2007)Google Scholar
- 4.Luštrek, M., Kaluža, B.: Fall detection and activity recognition with machine learning. Informatica 33, 205–212 (2008)Google Scholar
- 5.Albert, M.V., Kording, K., Herrmann, M., Jayaraman, A.: Fall classification by machine learning using mobile phones. PLoS ONE 7, 3–8 (2012)Google Scholar
- 8.Miaou, S.G., Sung, P.H., Huang, C.Y.: A customized human fall detection system using omni-camera images and personal information. In: Conference Proceedings - 1st Transdisciplinary Conference on Distributed Diagnosis and Home Healthcare, D2H2 2006, pp. 39–42 (2006)Google Scholar