Exploiting Physiological Sensors and Biosignal Processing to Enhance Monitoring Care in Mental Health

In this chapter, we describe how it is possible to exploit physiological sensors and related signal processing methods to enhance monitoring care mental health. Specifically, focusing on wearable sensors for Autonomic Nervous System (ANS) dynamics, we report on recent progresses in monitoring mood swings associated with bipolar disorder through the so-called PSYCHE system. Current clinical practice in diagnosing patients affected by this psychiatric disorder, in fact, is based only on verbal interviews and scores from specific questionnaires. Furthermore, no reliable and objective psycho- physiological markers are currently taken into account. We particularly describe a pervasive, wearable, and personalized system based on a comfortable t-shirt with integrated fabric electrodes and sensors able to acquire electrocardiogram, respirogram, and body posture information. In order to identify a pattern of objective physiological parameters to support the diagnosis, we describe ad-hoc methodologies of advanced biosignal processing able to effectively recognize four possible clinical mood states in bipolar patients (i.e., depression, mixed state, hypomania, and euthymia) who underwent long- term (up to 24h) monitoring. Mood assessment is here intended as an intra-subject evaluation in which the patient’s states are modeled as a stochastic process with time dependency, i.e., in the time domain, each mood state refers to the previous one(s). Experimental results are reported in terms of statistical analysis, as well as confusion matrices from automatic mood state recognition, and demonstrate that wearable and comfortable ANS monitoring could be a viable solution to enhance monitoring care in mental health. We conclude the chapter describing a methodology predicting mood changes in bipolar disorder using heartbeat nonlinear dynamics exclusively.