The long-term goal of our project is to develop a tool that can be used to diagnose and assess sleep disorders unobtrusively in a patient's home. Sleep disorders and sleep deprivation are significant public health problems. The U.S. Institute of Health estimates that 70 million Americans suffer from chronic, treatable sleep disorders. One of the most common and problematic sleep disorders is obstructive sleep apnea (OSA), where a partial collapse or obstruction of the pharyngeal airway results in intermittent reduction in blood oxygen saturation and sleep arousal. The traditional gold standard for diagnosing and monitoring these disorders is overnight polysomnography (PSG). Unfortunately PSG is expensive, obtrusive, and inconvenient. A less expensive tool that can be used to screen for SDB in a patient's home over multiple nights would help clinicians decide if polysomnography is indicated, and may provide important additional data about the nightly variance of the patient's sleep problems. In this study, we will develop algorithms for detecting SDB from data collected using load cells placed under the supports of the bed. Small movements in the body's center of mass allow breathing and movements to be detected as changes in the relative load at each corner of the bed. We hypothesize that these changes in load can be used to quantify the frequency and severity of apneas and hypopneas.
Our Specific Aims are: (1) to develop algorithms for detecting sleep disordered breathing and sleep efficiency from unobtrusive bed sensor data;(2) to determine if unobtrusive bed sensors can be used to classify the severity of sleep disordered breathing in a clinical population;and (3) to determine if unobtrusive bed sensors can be used in the home to screen for a positive apnea-hypopnea index (AHI).
Estimates of the prevalence of sleep disorders in the US range from 50 to 70 million people, and as many as 9% of middle-aged American men suffer from sleep disordered breathing. The direct cost of treating sleep disorders has been estimated in the range of $30-50 billion per year;indirect costs including absenteeism from work and fatigue-related accidents is estimated to be $210 billion. The proposed study would create a tool allowing unobtrusive in-home assessment of sleep-disordered breathing, for screening patients and following treatment.
| Mosquera-Lopez, Clara; Leitschuh, Joseph; Condon, John et al. (2018) In-Home Sleep Apnea Severity Classification using Contact-free Load Cells and an AdaBoosted Decision Tree Algorithm. Conf Proc IEEE Eng Med Biol Soc 2018:6044-6047 |
| Beattie, Zachary T; Jacobs, Peter G; Riley, Thomas C et al. (2015) A time-frequency respiration tracking system using non-contact bed sensors with harmonic artifact rejection. Conf Proc IEEE Eng Med Biol Soc 2015:8111-4 |
| Beattie, Zachary T; Hayes, Tamara L; Guilleminault, Christian et al. (2013) Accurate scoring of the apnea-hypopnea index using a simple non-contact breathing sensor. J Sleep Res 22:356-62 |
| Adami, Adriana M; Adami, André G; Hayes, Tamara L et al. (2012) A Gaussian model for movement detection during sleep. Conf Proc IEEE Eng Med Biol Soc 2012:2263-6 |
| Austin, Daniel; Beattie, Zachary T; Riley, Thomas et al. (2012) Unobtrusive classification of sleep and wakefulness using load cells under the bed. Conf Proc IEEE Eng Med Biol Soc 2012:5254-7 |
| Beattie, Zachary T; Hagen, Chad C; Hayes, Tamara L (2011) Classification of lying position using load cells under the bed. Conf Proc IEEE Eng Med Biol Soc 2011:474-7 |
| Adami, Adriana M; Adami, Andre G; Schwarz, Gilmar et al. (2010) A subject state detection approach to determine rest-activity patterns using load cells. Conf Proc IEEE Eng Med Biol Soc 2010:204-7 |
| Beattie, Zachary T; Hagen, Chad C; Pavel, Misha et al. (2009) Classification of breathing events using load cells under the bed. Conf Proc IEEE Eng Med Biol Soc 2009:3921-4 |
