Physiological Monitoring and Analysis

The physiological parameters were obtained through Empatica E4, a wearable device in the form of a wristband that allows for measuring Electrodermal Activity (EDA), blood volume pulse – from which heart rate (HR) and heart rate variability (HRV) are derived, skin temperature, and movement (Garbarino et al., 2014 ). EDA is a property of the skin that underlines the variation of the electrical conduction in response to sweat secretions and it is a sympathetic index (Boucsein, 2012 ). HRV is the physiological phenomenon of variation in the time interval between heartbeats. It is a parasympathetic index that reflects vagal activity, it is measured with the root mean square of the successive differences between inter beats intervals (Ernst, 2017 (link); Kim et al., 2018 (link)). Participants were asked to wear the Empatica E4 for 24 h on their non-dominant hand. HR was expressed in beats per minute (bpm) and derived through Empatica algorithms to the blood volume pulse. They provide also the inter beats intervals (IBI) from photoplethysmography (PPG) signal. HRV was obtained by extraction of the root mean square of successive differences between normal heartbeats (RMSSD) extracted by first calculating from IBI each successive time difference between heartbeats in ms, over a short-term period of 30 s. Then, each of the values was squared and averaged before the square root of the total was obtained. The sensor used to detect blood volume pulse is a PPG sensor, which is known to be subject to missing data as a result of movement or pressure artifacts (Chen et al., 2015 (link)). Artifacts were removed, discarding zero values and other single data point outliers. The analysis of EDA included the extraction of a parameter called skin conductance level (SCL). The electrodes used were silver coated with copper underlay on brass electrodes. The threshold for the amplitude of significant signal was set to a minimum rise of 0.005 μSiemens. SCL values were then normalized using the min-max method. Physiological data were down-sampled to 1hz and labeled as belonging to one of the three monitoring periods (defined as condition “work,” “sleep” and “daytime”). Data of each subject were then aggregated into periods, expressing their mean values.