作者: Abdul Q. Javaid , Carlo M. Noble , Russell Rosenberg , Mary Ann Weitnauer
关键词:
摘要: In this work, we apply machine learning to investigate the effectiveness of an Impulse Radio Ultra-Wide Band (IR-UWB) radar panel, in under-the-mattress configuration, for detecting apnea events subjects known have obstructive sleep (OSA). We consider a collection features, some novel and inspired by features that worked well detection using other types sensors (i.e., not IR-UWB). To extract collected total 25 hours data from four as they slept through night. The included digitized samples IR-UWB return signal scored polysomnograph (PSG), which is gold standard measures large number physiological parameters well-equipped laboratory. Normal epochs were extracted corresponding normal PSG data. Statistical derived these Linear Discriminant classifier was trained. Using cross-validation, found had accuracy around 70% epochs. aspect project involves processing investigation different methods feature extraction on obtained real suggests radar, when paired with might provide effective screening device convenient form factor.