Measurements of Sit-to-Stand Timing and Symmetry From Bed Pressure Sensors

Abstract

Sit-to-stand (SiSt) analysis has been widely used in clinical practice to assess the risk of falls in the older adult population. This paper proposes automated algorithms for the unobtrusive measurement of SiSt timing and symmetry using bed pressure sensors. An integrated signal comprising all of the sensor outputs was analyzed to measure both the bed-departure timing and the timing of three clinical phases within the transfer. Data collected in clinical trials, along with independent clinical video analysis, verified the success of the bed-departure timing algorithm with a mean error of 0.11 s. The phase measurement algorithm showed significant differences (p <; 0.001) between younger and older adults in Phases II and III of the transfers, comparing well with studies found in recent clinical literature. The sensor outputs were then used to form sequences of pressure images, and an automated region of interest (ROI) detection algorithm was designed to extract regional signals from the hips and the hands. The final algorithm was designed to measure the symmetry of the body throughout the SiSt transfer from the extracted regional signals. A system accuracy of 93.0% was obtained for the automated symmetry classification of transfers. The techniques proposed in this paper can increase the precision and efficiency in clinical SiSt assessments. Their unobtrusive nature makes them particularly suitable for integration into a continuous monitoring system such as those required within the smart home environment.