Content uploaded by Houtan Jebelli
Author content
All content in this area was uploaded by Houtan Jebelli on Mar 11, 2019
Content may be subject to copyright.
A preview of the PDF is not available
EEG-based workers' stress recognition at construction sites
Abstract and Figures
Taking into account that many construction workers suffer from excessive stress that adversely impacts their safety and health, early recognition of stress is an essential step toward stress management. In this regard, an electroencephalogram (EEG) has been widely applied to assess individuals' stress by analyzing brain waves in the clinical domains. With recent advancements in wearable EEG devices, EEG's ability can be extended to field workers, particularly by non-invasively assessing construction workers' stress. This study proposes a procedure to automatically recognize workers' stress in construction sites using EEG signals. Specifically, the authors collected construction field workers' EEG signals and preprocessed them to capture high-quality signals. Workers' salivary cortisol, a stress hormone, was also collected to label low or high-stress levels when they work at sites. Time and frequency domain features from EEG signals were calculated using fixed and sliding windowing approaches. Finally, the authors applied several supervised learning algorithms to recognize workers' stress while they are working at sites. The results showed that the fixed windowing approach and the Gaussian Support Vector Machine (SVM) yielded the highest classification accuracy of 80.32%, which is very promising given the similar accuracy of stress recognition in clinical domains where extricate and wired EEG devices were used and the subjects engage in minimal body movement. The results demonstrate that the proposed field stress recognition procedure can be used for the early detection of workers' stress, which can contribute to improving workers' safety, health, wellbeing, and productivity.
Figures - uploaded by Houtan Jebelli
Author content
All figure content in this area was uploaded by Houtan Jebelli
Content may be subject to copyright.
... Recently, Heart rate, heart rate variability and skin temperature were used as the physiological impact of the construction VR experience study [26]. "EEG is a noninvasive measurement of the brain's electrical activity, which is generated by firing neurons in the brain" [27]. EEG data is commonly used to examine two fundamental properties of affective experience: valence and arousal. ...
... Based on the famous active learning theory [30], creating excitement in the classroom can motivate students to be active learners. Recent years have seen a surge in the use of EEG signals in construction studies, primarily because wire-less wearable EEG devices have expanded its functionality non-intrusively, moreover, EEG offers rich information about an individual's mental status [27]. [31] compared three different light design alternatives by using VR simulation and the EEG of the participants was recorded to assess their emotions. ...
... To implement our approach, we use SVM classifier trained on a comprehensive feature set derived from public attention datasets. This classifier effectively blends linear and nonlinear methods with subjective analysis (Huang et al., 2023;Zhang et al., 2023), aiming to capture the complex dynamics of neural activity associated with attention (Jebelli et al., 2018). ...
Brain-computer interface (BCI) systems are highly valued for their applications in medicine and biology, where accurate analysis of attention levels and electroencephalography (EEG) data is crucial for the success rate of BCI systems. To improve this aspect, we propose to perform attention level determination prior to the execution of the BCI system. To accurately determine the level of attention, we propose a novel feature extraction approach that involves multidimensional feature extraction across multiple frequency bands of EEG data. The multi frequency band waveforms extracted using this method can be cross validated, thereby increasing the robustness of our research. We used publicly available datasets to train a Support Vector Machine (SVM) classifier to develop an efficient attention detection system, and developed a system for collecting EEG data to validate attention levels. The effectiveness and accuracy of the multidimensional feature extraction method was validated by classifying the data collected by the attention detection system. This study highlights the potential of integrating attention detection into BCI systems, pathways for advances in brain science research.
... Electroencephalography (EEG) presents a reliable and objective method for examining the neural correlates of specific cognitive processes through direct, and temporally precise, recording of the brain's electrical activity [27]. Past studies have used EEG as the main tool for measuring and monitoring neural activity in construction workers to evaluate different states such as mental workload [28][29][30], emotional state [31][32][33], fatigue [34][35][36][37][38] and stress [39,40]. ...
... The significance of EEG in studying construction safety issues is increasingly recognized in research. Jebelli et al. [13] developed a method for automatically identifying stress levels in construction workers by analyzing EEG signals, paving the way for early stress detection. Wang et al. [14] introduced a novel approach combining kinematic data with EEG to derive vigilance measurement indices, enhancing understanding of workers' risk perception and safety management on construction sites. ...
... According to the Federal Highway Administration (FHWA) work zone facts and statistics report, there was a significant increase in fatal crashes at work zones from 863 in 2020 to 956 in 2021 [1]. Research has shown that excessive stress in workers heightens the risk of errors, injuries, and various health issues among workers while concurrently being associated with reduced productivity levels [2]. In roadway work zones, lane closures, proximity to moving vehicles, night shifts, and the presence of construction vehicles and employees can all cause stress in workers. ...
... Studies examining the relationship between workers' fatigue and body stability have mainly focused on measuring accumulated fatigue rather than imposing quantitatively distinct levels of fatigue on workers. For example, Jabelli et al. [45] involve measuring cortisol levels to analyze the impact of fatigue on the walking stability of construction workers. It is necessary to apply various quantifiable levels of fatigue to individual workers. ...
In the construction industry, falls, slips, and trips (FST) account for 42.3% of all accidents. The primary cause of FST incidents is directly related to the deterioration of workers’ body stability. To prevent FST-related accidents, it is crucial to understand the interaction between physical fatigue and body stability in construction workers. Therefore, this study investigates the impact of fatigue on body stability in various construction site environments using Dynamic Time Warping (DTW) analysis. We conducted experiments reflecting six different fatigue levels and four environmental conditions. The analysis process involves comparing changes in DTW values derived from acceleration data obtained through wearable sensors across varying fatigue levels and construction environments. The results reveal the following changes in DTW values across different environments and fatigue levels: for non-obstacle, obstacle, water, and oil conditions, DTW values tend to increase as fatigue levels rise. In our experiments, we observed a significant decrease in body stability against external environments starting from fatigue Levels 3 or 4 (30% and 40% of the maximum failure point). In the non-obstacle condition, the DTW values were 9.4 at Level 0, 12.8 at Level 3, and 23.1 at Level 5. In contrast, for the oil condition, which exhibited the highest DTW values, the values were 10.5 at Level 0, 19.1 at Level 3, and 34.5 at Level 5. These experimental results confirm that the body stability of construction workers is influenced by both fatigue levels and external environmental conditions. Further analysis of recovery time, defined as the time it takes for body stability to return to its original level, revealed an increasing trend in recovery time as fatigue levels increased. This study quantitatively demonstrates through wearable sensor data that, as fatigue levels increase, workers experience decreased body stability and longer recovery times. The findings of this study can inform individual worker fatigue management in the future.
... Biofeedback can also be used in non-medical areas, which is the subject of scienti ic research [5][6][7][8][9][10][11][12]. It could be used to study the concentration of operators of unmanned ground vehicles (UGVs) or working machines. ...
Often, operators of machines, including unmanned ground vehicles (UGVs) or working machines, are forced to work in unfavourable conditions, e.g. high temperatures continuously for a long period of time. This has a huge impact on their concentration, which usually determines the success of many tasks entrusted to them. Electroencephalography (EEG) allows the study of the electrical activity of the brain. It allows determining, for example, whether the operator is able to focus on the realization of his tasks. The main goal of the article was to develop an algorithm for determining the state of brain activity by analysing the EEG signal. For this purpose, methods of EEG signal acquisition were described, basic types of brain waves were discussed, and exemplary states of brain activity were recorded. Particular attention was paid to technical aspects related to signal analysis. The LabVIEW environment was used to implement the created algorithm. The results of the research showing the operation of the developed EEG signal analyser were also presented.
Objective
This study aimed to determine the status of scientific production on biosensor usage for human health monitoring.
Methods
We used bibliometrics based on the data and metadata retrieved from the Web of Science between 2007 and 2022. Articles unrelated to health and medicine were excluded. The databases were processed using the VOSviewer software and auxiliary spreadsheets. Data extraction yielded 275 articles published in 161 journals, mainly concentrated on 13 journals and 881 keywords plus.
Results
The keywords plus of high occurrences were estimated at 27, with seven to 30 occurrences. From the 1595 identified authors, 125 were consistently connected in the coauthorship network in the total set and were grouped into nine clusters. Using Lotka's law, we identified 24 prolific authors, and Hirsch index analysis revealed that 45 articles were cited more than 45 times. Crosses were identified between 17 articles in the Hirsch index and 17 prolific authors, highlighting the presence of a large set of prolific authors from various interconnected clusters, a triad, and a solitary prolific author.
Conclusion
An exponential trend was observed in biosensor research for health monitoring, identifying areas of innovation, collaboration, and technological challenges that can guide future research on this topic.
Construction workers' emotional states (e.g., pleasure, displeasure, excitement, and relaxation) are known as a critical factor that affect their performance (e.g., safety, health, and productivity). To prevent adverse impacts on work performance, measuring emotional states should take precedence to better understand how workers' emotions vary while they are working. Among many methods available to measure emotional states, electroencephalogram (EEG) has a great potential for quantitative measurement by overcoming a possible bias from the survey-based subjective assessment of emotions. Although EEG-based emotion measurement has been tested and applied only in a laboratory environment, recent advancements in wearable EEG sensors, which are portable, wireless, and affordable, open a new door toward nonintrusive field emotion measurement. This study thus investigates the feasibility of measuring workers' emotions in the field using a wearable EEG sensor. To do this, a bipolar dimensional emotion model, which consists of valence (from displeasure to pleasure) and arousal (from relaxation to excitement) dimensions, was applied to quantify workers' emotional states. Then, workers' valence and arousal levels were measured using a wearable EEG sensor during their ongoing tasks. The validity of the EEG-based emotion measurement was examined through a comparison with cortisol levels obtained from workers' saliva samples, which has been accepted as a reliable physical measure of emotions. The results demonstrate the applicability of a wearable EEG sensor for measuring workers' emotions, particularly valence levels, which remain crucial to understanding workers' emotional states. This study contributes to the body of knowledge on in-depth studies for understanding workers' emotions in the field by providing a means to continuously and nonintrusively measure workers' emotions while they are working.
Poor work conditions and excessive physical demands associated with construction tasks increase the risks associated with construction workers’ impaired mental and physical health. To prevent detrimental impacts that such risks have on project performance, measuring construction workers’ physical and mental status should take precedence. In the same vein, recent advancements in wearable technologies and physiology have opened a new door toward an objective, nonintrusive, and continuous field measurement of workers’ physical and mental status using physiological signals acquired from wearable devices. Despite this opportunity, there is a notable paucity of studies investigating the potential of using physiological signals collected from wearable devices to understand workers’ physical and mental status. To bridge this gap, this study suggests utilizing an off-the-shelf wristband-type wearable sensor to acquire construction workers’ physiological signals as a means of assessing workers’ physical and mental state. This study investigates the distinguishing power of three biosignals—electrodermal (EDA), skin temperature (ST), and photoplethysmogram (PPG)—in detecting workers’ physical and mental states during their work on the site. The 10 h of biosignals were acquired from two workers in different conditions (e.g., light work, medium work) and tasks (e.g., resting, installing drywall, etc.). Signal processing methods (e.g., filtering methods) were applied to remove the signal artifacts from the biosignals. Electrodermal response (EDR), mean ST values after finite impulse responses filtering, heart rate variability (HRV), inter-beat interval (IBI), and percentage heart rate (%HRR) were calculated as the metrics to investigate the potential of suggested biosignals in measuring workers’ physical and mental states. Results indicated a clear difference in EDA values, HRV, and %HRR values while subjects were working in various real work conditions. Results confirm the feasibility of the wristband-type wearable sensor to evaluate construction workers’ physical and mental states, which can lead to early detection of the stressor factors in construction sites.
Construction is known as one of the most stressful occupations due to its involvement with physiologically and psychologically demanding tasks performed in a hazardous work environment. Because workers’ stress is a critical factor that adversely affects workers’ productivity, safety, well-being, and work quality, understanding workers’ stress should take precedence in the management of excessive stress. Various instruments for subjective measurement towards one’s perceived stress have been used, but such methods rely on imprecise memory and reconstruction of feelings in the past. Recent advancements in wearable Electroencephalography (EEG) devices possess a potential for quantitative measurement of human stress by directly capturing central nervous system activities from stress. However, its capability of measuring field workers’ stress under real occupational stressors remains questionable. This research thus proposes a framework to recognize construction workers’ stress at the field based on their brain activity recorded from a wearable EEG. Specifically, this framework applies a supervised learning algorithm—support vector machine—in detecting workers’ stress while working in different conditions. Workers salivary cortisol levels, which is a stress-related hormone, were used to label the tasks as low or high-stress level. Relevant time and frequency domain features in EEG signals were calculated. Results yielded a high of 71.1% accuracy using SVM in recognizing workers’ stress, which is a very promising result given that stress recognition with an exquisite EEG device in the clinical domain has at most the similar level of accuracy. The results show the potential for recognizing construction workers’ stress at the field by applying machine learning algorithms using workers’ brain waves recorded from a wearable EEG device. This EEG based stress detection approach will help to enhance workplace environment and conditions as well as to improve workers’ health by early detection and mitigation of the factors that cause stress.
Construction companies suffer huge losses due to labor fatalities and injuries. Since more than 70% of all accidents are related to human activities, detecting and mitigating human-related risks holds the key to improve the safety condition of construction industry. Many research reveals the psychological and emotional conditions of workers could contribute to the fatalities and injuries. More recent observations in the area of neural science and psychology suggest inattentional blindness is one major cause of unexpected human related accidents. Due to the limitation of human mental workload, labors are vulnerable to unexpected hazards while they are focusing on complicated construction tasks. Therefore, detecting the mental conditions of workers could indicate the hazards level of unexpected injuries. However, there is no available measurement can monitor construction workers’ mental condition and related hazards. This proposed research aims at proposing a measurement framework to evaluate such hazards through a neural time-frequency analysis approach. At the same time, the researchers also developed a prototype wearable Electroencephalography (EEG) safety helmet to enable the neural information collection.
Investigating brain waves collected by an electroencephalogram (EEG) can be useful in understanding human psychosocial aspects such as stress, emotional exhaustion, burnout, and mental fatigue. Recently, an off-the-shelf wearable EEG device, which is wireless, lightweight, and affordable, has become available so that field construction workers’ psychosocial status can be explored without interfering their ongoing work. However, capturing high quality EEG signals from such a device can be very challenging on real construction sites due to the signal artifacts generated from body movement caused by physically demanding work. To address this issue, the authors propose an EEG signal-processing framework that can acquire high quality EEG signals from real construction sites using a wearable EEG device. Specifically, the signal-processing framework reduces noises and is thus enabled to extract quality EEG signals. This framework is validated by examining whether the brain activation (particularly, brain activation by body movements) can be identified using the processed EEG signal applied to eight field construction workers’ under working (i.e., active) and not working (i.e., inactive) conditions. Specifically, mean power spectral density (PSD) of the EEG beta frequency range is calculated from electrodes near the motor cortex of the brain that controls voluntary movements. A significant difference in mean PSD in the beta frequency range between active and inactive conditions demonstrates that the processed EEG signal, based on the proposed framework, captures brain activation. The results show the potential of the proposed signal processing framework to monitor workers’ brain wave patterns on the field with a wearable EEG device, which opens a door to assess workers’ psychosocial status in construction so that any psychosocial problems of workers can be investigated.
This paper presents an investigation about the effects of mental stress on prefrontal cortex (PFC) subregions using simultaneous measurement of functional Near-Infrared Spectroscopy (fNIRS) and Electroencephalography (EEG) signals. The aim is to explore canonical correlation analysis (CCA) technique to study the relationship among the bi-modality signals in mental stress assessment, and how we could fuse the signals for better accuracy in stress detection. Twenty-five male healthy subjects participated in the study while performing mental arithmetic task under control and stress (under time pressure with negative feedback) conditions. The fusion of brain signals acquired by fNIRS-EEG was performed at feature-level using CCA by maximizing the inter-subject covariance across modalities. The CCA result discovered the associations across the modalities and estimated the components responsible for these associations. The experiment results showed that mental stress experienced by this cohort of subjects is subregion specific and localized to the right ventrolateral PFC subregion. These suggest the right ventrolateral PFC as a suitable candidate region to extract biomarkers as performance indicators of neurofeedback training in stress coping.
Inappropriate task allocation in construction activities results not only in lower productivity but also higher error rates and more work-related injuries. Without quantitative assessment of task demand, it is nearly impossible to achieve reasonable and flexible task allocation. However, it is extremely difficult to assess the mental/cognitive demand of construction tasks quantitatively and reliably. This paper introduces a novel electroencephalography (EEG) approach to estimate task mental workload based on the power spectral densities (PSDs) of major frequency bands. A preliminary experiment was designed and conducted to demonstrate the proposed assessment approach, and the National Aeronautics and Space Administration task load index (NASA-TLX) was adopted to validate the results. Four EEG channels were tested for 30 subjects in the preliminary experiment with three-way analysis of variance (ANOVA). Both the types of construction activities and frequency bands show statistically significant effects on PSDs of the EEG signals. Among all channels, the channel that is located at the left frontal parietal lobe (Fp1), especially the gamma band, showed the highest correlation with the mental load level. The final statistical results and the frequency bin model suggest that the estimation results are consistent with the NASA-TLX mental demand scores.
This paper presents an investigation about the effects of mental stress on prefrontal
cortex (PFC) subregions using simultaneous measurement of functional Near-Infrared
Spectroscopy (fNIRS) and Electroencephalography (EEG) signals. The aim is to explore
canonical correlation analysis (CCA) technique to study the relationship among the bimodality signals in mental stress assessment, and how we could fuse the signals for better
accuracy in stress detection. Twenty-five male healthy subjects participated in the study while
performing mental arithmetic task under control and stress (under time pressure with negative
feedback) conditions. The fusion of brain signals acquired by fNIRS-EEG was performed at
feature-level using CCA by maximizing the inter-subject covariance across modalities. The
CCA result discovered the associations across the modalities and estimated the components
responsible for these associations. The experiment results showed that mental stress
experienced by this cohort of subjects is subregion specific and localized to the right
ventrolateral PFC subregion. These suggest the right ventrolateral PFC as a suitable candidate
region to extract biomarkers as performance indicators of neurofeedback training in stress
coping.
As a sector associated with high injury and fatality rates, the construction industry requires constant caution with regard to construction laborers during project execution. Different from people in other industries, construction workers are less sensitive to hazards because of their long-term exposure to risks. Therefore, maintaining construction workers' vigilance and monitoring their attention levels are critical to successful safety management practices. However, current attention-assessing approaches are post hoc and subjective and difficult to implement in construction practice. To address these issues, we propose a wireless and wearable electroencephalography (EEG) system to quantitatively and automatically assess construction workers' attention level through processing human brain signals. To validate the proposed system, we conducted an on-site experiment to analyze the EEG signal patterns when construction workers avoid different obstacles in their tasks. The results suggest EEG signal properties such as frequency, power spectrum density, and spatial distribution can effectively reflect and quantify the construction workers' perceived risk level. Especially, lower gamma frequency bands and the frontal left EEG cluster provide the most direct and observable indications of their vigilance states. These conclusions could facilitate the future implementation of wearable EEG devices through data filtering and channel optimization.