Monitoring of patients suffering from REM sleep behavior disorder

Rapid eye movement (REM) sleep behavior disorder (RBD) is a parasomnia that involves involuntary, unwanted, and random movements of a dreaming patient. Typically, these dreams contain violent activities. There is a high likelihood of the patient being injured or hurting his bed-partner as a result of these enactments. Continuous monitoring of sleeping RBD patients can prevent these harmful events through timely intervention. This paper presents a novel method for continuous observation of RBD patients exploiting fine-grained amplitude and phase information of the wireless channel response. The variations in the wireless channel response as a result of different patient movements are assessed and used to identify RBD episodes. The data obtained are classified using a support vector machine and deliver an accuracy level of more than 90%. To the best of authors' knowledge, this is a first attempt at using radio frequency signals to sense RBD in real time.