Physiological signal based work stress detection using unobtrusive sensors

Abstract

Objective: Work stress is identified as the 'health epidemic of 21st century' by WHO because, when left unchecked, it wreaks havoc on human mind and body by accelerating the onset and progression of several health disorders. Hence, the evolution of strategies for early detection of mental stress is pivotal. The study presented here is one step towards the goal of developing a physiological parameter based psychological stress detection scheme which can further be incorporated into a wearable vital signs monitor. Approach: A group of 34 subjects (14 females and 20 males, age: 21.4 ± 1.7 years; mean ± SD) volunteered to participate in a pilot laboratory intervention that emulated real-life job stress scenarios by incorporating stress factors like mental workload, time pressure, performance pressure and social evaluative threat. Electrodermal Activity (EDA), Electrocardiogram (ECG), and Skin Temperature (ST) were monitored throughout the experiment to capture sympathetic activation during stress. Stress response elicitation was validated using salivary cortisol levels. A total of 61 features were extracted from these signals and four classifiers were investigated regarding their ability to detect 'stress' using single and multimodal schemes. A fusion framework that combined the benefits of feature fusion and decision fusion was employed to generate classifier ensembles for multimodal stress detection schemes. As the generated datasets exhibited a class imbalance issue, three separate schemes for class imbalance rectification viz., undersampling, oversampling and SMOTE were investigated concerning their ability to yield the best classification performance. While ECG based performance analysis was restricted to data segments of 300 s duration to conform to international guidelines for short-term HRV analysis, non-overlapping EDA and ST data segments of durations 300 s, 180 s, 60 s, and 30 s were examined to determine the optimum data length that can generate best results. Main Results: EDA gave a superior performance for 60 s windows while ST performed best with data segments of duration 30 s. A comparative study was performed with 25%, 50%, 75% and 90% overlapping data segments as well. However, overlapping did not enhance the performance of the classifiers significantly.While EDA emerged as the best single modality, the highest stress recognition accuracy of 97.13% was yielded by a combination of EDA and ST with data segments of 60 s duration. Furthermore, the differential effect of 'physical' and 'psychological' stressors on EDA and ST was analyzed. Significance: The results clearly suggest that these physiological parameters can not only reliably detect psychological stress but can also discriminate it from physical stress.

Multimodal Features for Detection of Driver Stress and Fatigue: Review

“…This group includes the following advanced features: sample entropy [21]; detrended fluctuation analysis DFA-α1 and DFA-α2 [21]; STD of the short-term and long-term RR interval variability (SD1, SD2, respectively) derived from the Poincaré plots [21]. HR and HRV are one of the most significant indicators of stress [27], [57], [76]. The original signal is inexpensive and not difficult to record [75] and can be relatively easily measured inside the cockpit.…”

“…Among others, two features were extracted from EDA signal in 10 s window in [22]: mean and standard deviation for stress, fatigue, drowsiness and normal state distinction. In [76] the EDA signal was used together with temperature and ECG filtered signals for stress detection (not in driver). EDA signal was filtered using Butterworth low pass filter with cut-off frequency of 5 Hz.…”

“…Phasic component was calculated as a difference between filtered EDA and tonic component. From filtered EDA signal, tonic and phasic components several features were calculated, namely: mean, variance, difference between maximum and minimum (range), ratio of range to absolute mean, ratio of standard deviation to mean and thresholds for which 10%, 25%, 50%, 75% and 90% of data values are smaller than these thresholds [76].…”

“…Anusha et al [76] extracted seven different thermal features based on evaluation of thermograms of finger and wrist: mean temperature, minimum and maximum of temperature, its standard deviation, difference between maximum and minimum of temperatures (range), ratio of range to absolute mean and ratio of standard deviation to mean. These features were used with features from ECG and EDA (total of 61 features) for detection of psychological stress.…”

Personal Productivity Management in the Digital Age: Measures from Research and Use of Conventional Tools

“…While this parameter no longer appeared in the analyzed articles in the following years, it was increasingly used again in 2018. The skin temperature was also examined as a possible parameter in these publications for the first time that year [13,14].…”