Towards unobtrusive automated sleep stage classification: polysomnography using electrodes on the face

Although sleep stage annotation (SSA) is historically known from clinical practice and typically performed by a certified expert on the basis of visual examination of polysomnography (PSG) signals. Automatic SSA has emerged as a tool to assist sleep experts and to accelerate the analysis of PSG data. New advances in signal processing and sensor technology start to enable the application of SSA in home solutions as well. In today's busy lives, sleep plays a central role and good quality sleep helps us to deal with the stress of everyday life. Being able to enhance sleep quality thus is a major opportunity to help people in reducing the influence of stress on their live, health and wellbeing. The advent of consumer products aimed at enhancing the sleep experience has propelled the need for home sleep monitoring and inducing solutions which can i) provide automatic SSA using sensors that interfere minimally with the sleep process and ii) provide sleep stage information in real-time in order to be suitable for closed-loop sleep inducing solutions. In this paper, we examine two possible alternatives for unobtrusive sleep monitoring. The first one uses respiratory, cardiac and wrist actigraphy signals while the second one relies on Facial PSG electrodes positioned on the facial area which allow for unobtrusive and comfortable sensors arrangements.