Wearable Biosensors to Understand Construction Workers' Mental and Physical Stress
dc.contributor.author | Jebelli, Houtan | |
dc.date.accessioned | 2019-07-08T19:45:22Z | |
dc.date.available | NO_RESTRICTION | |
dc.date.available | 2019-07-08T19:45:22Z | |
dc.date.issued | 2019 | |
dc.date.submitted | 2019 | |
dc.identifier.uri | https://hdl.handle.net/2027.42/149965 | |
dc.description.abstract | Occupational stress is defined as harmful physical and mental responses when job requirements are greater than a worker's capacity. Construction is one of the most stressful occupations because it involves physiologically and psychologically demanding tasks performed in a hazardous environment this stress can jeopardize construction safety, health, and productivity. Various instruments, such as surveys and interviews, have been used for measuring workers’ perceived mental and physical stress. However valuable, such instruments are limited by their invasiveness, which prevents them from being used for continuous stress monitoring. The recent advancement of wearable biosensors has opened a new door toward the non-invasive collection of a field worker’s physiological signals that can be used to assess their mental and physical status. Despite these advancements, challenges remain: acquiring physiological signals from wearable biosensors can be easily contaminated from diverse sources of signal noise. Further, the potential of these devices to assess field workers’ mental and physical status has not been examined in the naturalistic work environment. To address these issues, this research aims to propose and validate a comprehensive and efficient stress-measurement framework that recognizes workers mental and physical stress in a naturalistic environment. The focus of this research is on two wearable biosensors. First, a wearable EEG headset, which is a direct measurement of brain waves with the minimal time lag, but it is highly vulnerable to various artifacts. Second, a very convenient wristband-type biosensor, which may be used as a means for assessing both mental and physical stress, but there is a time lag between when subjects are exposed to stressors and when their physiological signals change. To achieve this goal, five interrelated and interdisciplinary studies were performed to; 1) acquire high-quality EEG signals from the job site; 2) assess construction workers’ emotion by measuring the valence and arousal level by analyzing the patterns of construction workers’ brainwaves; 3) recognize mental stress in the field based on brain activities by applying supervised-learning algorithms;4) recognize real-time mental stress by applying Online Multi-Task Learning (OMTL) algorithms; and 5) assess workers’ mental and physical stress using signals collected from a wristband biosensor. To examine the performance of the proposed framework, we collected physiological signals from 21 workers at five job sites. Results yielded a high of 80.13% mental stress-recognition accuracy using an EEG headset and 90.00% physical stress-recognition accuracy using a wristband sensor. These results are promising given that stress recognition with wired physiological devices within a controlled lab setting in the clinical domain has, at best, a similar level of accuracy. The proposed wearable biosensor-based, stress-recognition framework is expected to help us better understand workplace stressors and improve worker safety, health, and productivity through early detection and mitigation of stress at human-centered, smart and connected construction sites. | |
dc.language.iso | en_US | |
dc.subject | occupational stress | |
dc.subject | wearable biosensors | |
dc.subject | stress-measurement framework | |
dc.subject | worker safety, health, and productivity | |
dc.subject | machine learning | |
dc.subject | Electrodermal activity [EDA], Electroencephalography [EEG], Photoplethysmography [PPG], Electrocardiography [ECG], Electromyographm (EMG) | |
dc.title | Wearable Biosensors to Understand Construction Workers' Mental and Physical Stress | |
dc.type | Thesis | |
dc.description.thesisdegreename | PhD | en_US |
dc.description.thesisdegreediscipline | Civil Engineering | |
dc.description.thesisdegreegrantor | University of Michigan, Horace H. Rackham School of Graduate Studies | |
dc.contributor.committeemember | Lee, SangHyun | |
dc.contributor.committeemember | Scott, Clayton D | |
dc.contributor.committeemember | Kamat, Vineet Rajendra | |
dc.contributor.committeemember | Menassa, Carol C | |
dc.subject.hlbsecondlevel | Civil and Environmental Engineering | |
dc.subject.hlbtoplevel | Engineering | |
dc.description.bitstreamurl | https://deepblue.lib.umich.edu/bitstream/2027.42/149965/1/hjebelli_1.pdf | |
dc.identifier.orcid | 0000-0003-4786-7616 | |
dc.identifier.name-orcid | Jebelli, Houtan; 0000-0003-4786-7616 | en_US |
dc.owningcollname | Dissertations and Theses (Ph.D. and Master's) |
Files in this item
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.