No abstract

The problem of stress detection and classification has attracted a lot of attention in the past decade. It has been tackled with mainly two different approaches, where signals were either collected in ambulatory settings, which can be limited to the period of presence in the hospital, or in continuous mode in the field. A sensor-based continuous measurement of stress in daily life has a potential to increase awareness of patterns of stress occurrence. In this work, we first present a data-flow infrastructure suitable for two types of studies that conforms with the data protection requirements of the ethics committee monitoring the research on humans. The detection and binary classification of stress events is compared with three different machine learning models based on the features (meta-data) extracted from physiological signals acquired in laboratory conditions and ground-truth stress level information provided by the subjects themselves via questionnaires associated with these features. The main signals considered in current classification are electro-dermal activity (EDA) and blood volume pulse (BVP) signals. Different models are compared and the best configuration yields an F1 score of 0.71 (random baseline: 0.48). The importance on prediction of phasic and tonic EDA components is also investigated. Our results also pave the way for further work on this topic with both machine learning approaches and signal processing directions.

No abstract

No abstract

In this work we present a system that integrates a wearable sensor equipped bracelet, a mobile application and a cloud-based back-end application that allows non-invasive monitoring of sleep patterns. We further analyzed a set of physiological signals and their features to explore which features among them indicate significant differences between sleep and wakefulness. Our study is fairly comprehensive with regard to the variety of physiological signals that we collect in ambulatory settings without disturbing the patient. In our analysis various signals such as electrodermal activity, heart rate variability and blood volume pulse and patients' movements registered via an accelerometer are taken into consideration. We show that by carefully estimating a rich set of parameters for support vector machine based classification algorithm we can achieve up to 93 % of correct classification when classifying sleep and wakefulness periods for the analyzed subjects. Our study shows that such a system can be of great use for medical health practitioners who are interested to follow sleep patterns and regularity for patients whose condition is at risk either during sleep or due to lack of sleep. In the future we aim to investigate more in details the different types and phases of sleep such as nREM1, nREM2, slow-wave and rapid-eye movement REM. For this purpose, we plan to compare a completely non-invasive methods such as the method proposed here, with hospital grade quality sleep evaluation methods. Integrating user feedback about the mobile application is also planned as a part of our future work.