IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI ...) · 2014 · 57 citations · 19 references
NursingHealth SciencesPsychophysiologySleep MedicineStress PhysiologySleep AssessmentSleep HealthSleep PhysiologySleep Signal AnalysisStress BiomarkersWearable Health TechnologiesBody TemperatureClassification ModelsMonitoring Sleep DurationSleepStress DetectionStress MonitoringWearable DevicesWearable Sensors
This work presents an approach for detecting pilgrims' stress levels using their nightly sleep patterns and identifying the most relevant sleep parameters for stress detection. We collected data from 10 participants, who wore two wearable devices, a chest strap and a wristband, during the 2013 Hajj pilgrimage. To build and evaluate different classification models we use bio-physiological measures such as ECG/HRV, respiration, body temperature and GSR data, and physical parameters including upper body posture sensors and accelerometers on the body and arms. The best classifier is capable of differentiating between low, moderate, and high-perceived stress with an accuracy of 73 %. Both physical and bio-physiological feature modalities show similar individual performance. Sleep duration and upper body activity on the one side and ECG/HRV features on the other are the most important sleep parameters.
19
A Global Measure of Perceived Stress
Sheldon Cohen, Robin Mermelstein · Journal of Health and Social Behavior · 1983 · 32.4K citations
Chronic stress at work and the metabolic syndrome: prospective study
Tarani Chandola, Eric J. Brunner, Michael Marmot · BMJ · 2006 · 1.1K citations · Full text
Hong Lu, Denise Frauendorfer, Mashfiqui Rabbi et al. · 2012 · 516 citations · Full text