A Sensitivity Analysis of Biophysiological Responses of Stress for Wearable Sensors in Connected Health

Abstract

Stress is known as a silent killer that contributes to several life-threatening health conditions such as high blood pressure, heart disease, and diabetes. The current standard for stress evaluation is based on self-reported questionnaires and standardized stress scores. There is no gold standard to independently evaluate stress levels despite the availability of numerous biophysiological stress indicators. With an increasing interest in wearable health monitoring in recent years, several studies have explored the potential of various biophysiological indicators of stress for this purpose. However, there is no clear understanding of the relative sensitivity and specificity of these stress-related biophysiological indicators of stress in the literature. Hence this study aims to perform statistical analysis and classification modelling of biophysiological data gathered from healthy individuals, undergoing various induced emotional states, and to assess the relative sensitivity and specificity of common biophysiological indicators of stress. In this paper, several frequently used key indicators of stress, such as heart rate, respiratory rate, skin conductance, RR interval, heart rate variability in the electrocardiogram, and muscle activation measured by electromyography, are evaluated based on a detailed statistical analysis of the data gathered from an already existing, publicly available WESAD (Wearable Stress and Affect Detection) dataset. Respiratory rate and heart rate were the two best features for distinguishing between stressed and unstressed states.

Selection of Optimal Physiological Features for Accurate Detection of Stress

“…Similar to other studies [13][14] [15], that have reported previously we found that EDA, temperature, and accelerometer are important features in stress detection. Our objective was not to find one optimal feature, but multiple features that can help in precise detection.…”

Stress Monitoring Using Wearable Sensors: A Pilot Study and Stress-Predict Dataset

“…Previously, in a detailed literature survey and statistical analysis to determine the most sensitive and specific parameters for stress-monitoring, we concluded that the respiratory rate (RR) is the most important parameter for the detection of stress conditions [ 15 , 16 , 33 ]. The results of these statistical analyses were published in [ 15 ].…”

“…Moreover, the lack of gold standard ground-truth/reference values or data, collection of stress data in the natural environment, different confounding variables, identification of discriminative/specific stress features, and development of an accurate classifier model to classify stress data from baseline/normal are also contributing reasons to the lack of unswerving stress monitoring device. Further details are explained in [ 15 ].…”

“…Different studies have used the same physiological variables and classifiers but have reported significantly different classification accuracies. Furthermore, there is no clear understanding of the relative sensitivity and specificity of stress-related biophysiological indicators of stress (such as heart rate and respiratory rate) in the literature [ 15 , 23 ]. All the above-mentioned datasets have a relatively small sample size and are more focused on performing classification analysis with reported accuracies in the range of from 60 to 70% rather than performing a statistical analysis of the dataset.…”

Exploring Unsupervised Machine Learning Classification Methods for Physiological Stress Detection

“…The physiological parameters that are frequently used for stress analysis are respiratory rate, heart rate, skin conductance, skin temperature, and galvanic skin response ( 13 ). As supervised learning requires training labels for training the classifier, in most cases, either the labels are unavailable or inaccurate, in the real-time data collection ( 14 ).…”