Dept. of Biomedical Engineering
Oregon Health & Science University
The quality of sleep is an important attribute of an individualâs health state and its assessment is therefore a useful diagnostic feature. Changes in the sleep-related behaviors as well as in the patterns of motor activities during sleep can be a disease marker, or can reflect various abnormal physiological and neurological conditions. Presently, there are no convenient, unobtrusive ways to assess and quantify the quality of sleep at point of care outside of a clinic. This dissertation describes an approach and a system for unobtrusive assessment of activity patterns and movement in bed that uses load cells installed at the corners of a bed. The system focuses on identifying when a movement occurs and on determining the type of movement performed based on the forces sensed by the load cells. The feasibility and accuracy of the movement detection and classification is evaluated using data collected in the laboratory and in a study with residents of an assisted-living facility (Elite Care, Milwaukie, OR). The movement detection approach estimates the energy in each load cell signal over short segments to capture the variations caused by movement. The average equal error rate of the detector is 3.22% (Â± 0.54). The performance of the detector is invariant with respect to the individualâs characteristics, e.g., weight, as well as those of the bed. The dissertation describes several approaches to signal representation and discrimination techniques for clinically relevant classification of the type of movements with the goal of weight-invariant performance. The results of correct classification for an approach based on Gaussian Mixture Models using a time-domain representation and a wavelet-based time-frequency representation, as evaluated by laboratory experiments, are 84.6% and 82.2%, respectively. The simplicity of the resulting algorithms, the relative insensitivity to the weight and height of the monitored individual and the minimal training requirements make the resulting approaches practical and easily deployable in residential and clinical settings.
OGI School of Science and Engineering
Adami, Adriana Miorelli, "Assessment and classification of movements in bed using unobtrusive sensors" (2006). Scholar Archive. 28.