Stress Monitoring in Construction

Stress Monitoring in Construction

• This presentation focuses on stress monitoring in construction, specifically using wearable technology.
• Interventions and management strategies are not covered; these are easily searchable.
• The presenter emphasizes the need for stress monitoring, even when typical stressors are known.
• Proactive management of worker interactions and site environments is crucial for safety, health, productivity, and job satisfaction.

Traditional Stress Monitoring Techniques

• Surveys are a common approach involving questionnaires.
• Perceived Stress Scale (PSS), Job Stress Survey (JSS), Stress Appraisal Measure (SAM), and Cognitive Appraisal Ratio are examples of survey methods.
• These methods often focus on a 30-day period and are usually specific to the workplace.
• Interviews and manual observations provide another way of gathering data, though these methods can be invasive and require significant time commitment.
• Traditional methods are not ideal for proactive interventions due to their invasive nature and lack of continuous data collection capabilities. These methods are better suited for post-hoc analysis.

Wearable Biosensors

• Traditional monitoring techniques have limitations, motivating investigation into wearable biosensors.
• Wearable biosensors offer potential advantages for continuous stress monitoring.
• These devices collect various biosignals from peripheral and central systems.

Different Biosignals Collected by Wearables

• ECG (Electrocardiogram): Measures the electrical activity of the heart.
• PPG (Photoplethysmogram): Measures changes in blood volume in response to light.
• EDA (Electrodermal Activity): Measures skin conductance, reflecting stress levels.
• ST (Skin Temperature): Measures skin temperature changes, which might be related to stress responses.
• Eye-tracking: Monitors eye movements, considered a potential indicator of stress.
• EEG (Electroencephalogram): Monitors the electrical activity in the brain.
• fNIRS (Functional Near-Infrared Spectroscopy): Measures brain activity using near-infrared light.

Electrocardiogram (ECG)

• ECG uses electrodes to measure heart activity which is linked to sympathetic arousal.
• Heart rate (HR), heart rate variability metrics (HRV), SDNN, RMSSD are key metrics used in ECG analysis and provide ways to measure stress responses (high heart rates increase under stress).
• Time and frequency domains are used to analyze ECG signals.

Photoplethysmogram (PPG)

• PPG measures changes in blood volume using light.
• Peaks in PPG correlate with R peaks in ECG helping in measuring HR and HRV metrics.
• PPG is less invasive and widely used, particularly with wearable technology.

Electrodermal Activity (EDA)

• EDA measures skin conductance, which is affected by stress.
• Tonic EDA measures overall stress levels, and phasic EDA responds to immediate stressors.
• Time and frequency domains provide analysis methods.
• EDA is reasonably affordable and provides information on both long-term and immediate stress responses.

Electroencephalogram (EEG)

• EEG measures brain wave activity and gives an indication of stress levels in the form of associated activity patterns.
• EEG can differentiate stress from threats and challenges; different frequency bands (Delta, Theta, Alpha, Beta, Gamma) are associated with various cognitive states, including stress responses.
• EEG analysis can be done network-wise and helps in understanding the cognitive process by focusing on active networks such as (DMN, CEN, SN) and microstates.

Wearable Biosensor + Machine-learning

• This presentation highlights the potential for using wearable biosensors combined with machine learning for stress monitoring in field settings.

Potential Barriers

• Motion artifact can negatively impact the accuracy of wearable biosensors.
• Environmental factors like humidity can affect the readings.
• Some methods are invasive, requiring user-interaction.
• Wearable technologies can be expensive.