D C Ranasinghe1, R L Shinmoto Torres, K Hill, R Visvanathan. 1. School of Computer Science, University of Adelaide, Australia; Auto-ID Lab, Faculty of Engineering Computer and Mathematical Sciences, University of Adelaide, SA 5000, Australia. Electronic address: damith.ranasinghe@adelaide.edu.au.
Abstract
INTRODUCTION: Falls in hospitals and residential care facilities commonly occur near the bed. The aim of this study was to investigate the accuracy of a continuously wearable, batteryless, low power and low cost monitoring device (Wearable Wireless Identification and Sensing Platform) with a single kinematic sensor capable of real-time monitoring to automatically detect bed entry and exit events. MATERIALS AND METHODS: Three dimensional acceleration readings and the strength of the transmitted signal from the WISP was interpreted to identify bed exit events and sensitivity, specificity and Receiving Operator Curves (ROC) were determined. RESULTS: The sensor located over sternum method performed best with sensitivity and specificity values of 92.8% and 97.5% respectively for detecting bed entry and values of 90.4% and 93.80% respectively for bed exit. On the other hand, the sensor-on-mattress algorithm achieved sensitivity and specificity values of 84.2% and 97.4% respectively for bed entry and 79% and 97.4% for bed exit detection. CONCLUSION: The WISP located over the sternum method is the preferred method to detect bed entry and exit. However, further work in frail older people is required to confirm the performance of this method.
INTRODUCTION: Falls in hospitals and residential care facilities commonly occur near the bed. The aim of this study was to investigate the accuracy of a continuously wearable, batteryless, low power and low cost monitoring device (Wearable Wireless Identification and Sensing Platform) with a single kinematic sensor capable of real-time monitoring to automatically detect bed entry and exit events. MATERIALS AND METHODS: Three dimensional acceleration readings and the strength of the transmitted signal from the WISP was interpreted to identify bed exit events and sensitivity, specificity and Receiving Operator Curves (ROC) were determined. RESULTS: The sensor located over sternum method performed best with sensitivity and specificity values of 92.8% and 97.5% respectively for detecting bed entry and values of 90.4% and 93.80% respectively for bed exit. On the other hand, the sensor-on-mattress algorithm achieved sensitivity and specificity values of 84.2% and 97.4% respectively for bed entry and 79% and 97.4% for bed exit detection. CONCLUSION: The WISP located over the sternum method is the preferred method to detect bed entry and exit. However, further work in frail older people is required to confirm the performance of this method.
Authors: Roberto Luis Shinmoto Torres; Renuka Visvanathan; Stephen Hoskins; Anton van den Hengel; Damith C Ranasinghe Journal: Sensors (Basel) Date: 2016-04-15 Impact factor: 3.576
Authors: Anna L Barker; Renata T Morello; Rory Wolfe; Caroline A Brand; Terry P Haines; Keith D Hill; Sandra G Brauer; Mari Botti; Robert G Cumming; Patricia M Livingston; Catherine Sherrington; Silva Zavarsek; Richard I Lindley; Jeannette Kamar Journal: BMJ Date: 2016-01-26
Authors: Roberto L Shinmoto Torres; Renuka Visvanathan; Derek Abbott; Keith D Hill; Damith C Ranasinghe Journal: PLoS One Date: 2017-10-09 Impact factor: 3.240
Authors: Renuka Visvanathan; Damith C Ranasinghe; Kylie Lange; Anne Wilson; Joanne Dollard; Eileen Boyle; Katherine Jones; Michael Chesser; Katharine Ingram; Stephen Hoskins; Clarabelle Pham; Jonathan Karnon; Keith D Hill Journal: J Gerontol A Biol Sci Med Sci Date: 2022-01-07 Impact factor: 6.053