Continuous Stress Detection Using Wearable Sensors in Real Life: Algorithmic Programming Contest Case Study

Abstract

The negative effects of mental stress on human health has been known for decades. High-level stress must be detected at early stages to prevent these negative effects. After the emergence of wearable devices that could be part of our lives, researchers have started detecting extreme stress of individuals with them during daily routines. Initial experiments were performed in laboratory environments and recently a number of works took a step outside the laboratory environment to the real-life. We developed an automatic stress detection system using physiological signals obtained from unobtrusive smart wearable devices which can be carried during the daily life routines of individuals. This system has modality-specific artifact removal and feature extraction methods for real-life conditions. We further tested our system in a real-life setting with collected physiological data from 21 participants of an algorithmic programming contest for nine days. This event had lectures, contests as well as free time. By using heart activity, skin conductance and accelerometer signals, we successfully discriminated contest stress, relatively higher cognitive load (lecture) and relaxed time activities by using different machine learning methods.

Keywords: daily life psychophysiological data; electrodermal activity; heart rate variability; machine learning; photoplethysmography; smartwatch; stress recognition; wearable sensors.

Figure 1

Recorded physiological signals before and after the start of the stimuli. The increase in EDA signal level and number of peaks and irregularities and sudden increases in HRV can be seen in this figure.

Figure 2

The block diagram of the stress level detection system for Samsung Gear S and S2 and Empatica E4. Since the sensors and platforms are different, please note that EDA and temperature signals are only available for E4.

Figure 3

The example filtered EDA signal according to changes in the accelerometer signal. Note that red components were deleted because of the high activity intensity.

Figure 4

Activity intensity is shown by using the accelerometer sensor X, Y, and Z components corresponding to the example EDA signal in Figure 3 Note that this example was recorded during a highly intensive activity.

Figure 5

Decomposed EDA Signal from Empatica E4 wristband by applying cvxEDA tool.

Figure 6

Gaps due to movement and loosely worn wristband from PPG (Photoplethysmography) data (Left) are filled with cubic interpolation function (Right).

Figure 7

A view of smartwatches and wristbands after data extraction, charged and ready to use.

Figure 8

The daily schedule and data collection procedure during the algorithmic programming contest.

Figure 9

Percentage of the remaining data (for both device types) after the artifacts are removed versus different percentage thresholds of artifact detection.