Houtan Jebelli

Comprehensive Fall-Risk Assessment of Construction Workers Using Inertial Measurement Units: Validation of the Gait-Stability Metric to Assess the Fall Risk of Iron Workers

AbstractIn construction worksites, slips, trips, and falls are major causes of fatal injuries. This fact demonstrates the need for a safety assessment method that provides a comprehensive fall-risk analysis inclusive of the effects of physiological characteristics of construction workers. In this context, this research tests the usefulness of the maximum Lyapunov exponents (Max LE) as a metric to assess construction workers’ comprehensive fall risk. Max LE, one of the gait-stability metrics established in clinical settings, estimates how the stability of a construction worker reacts to very small disruptions. In order to validate the use of Max LE, a laboratory experiment that asked a group of subjects to simulate iron workers’ walking tasks on an I-beam was designed and conducted. These tasks were designed to showcase various fall-risk profiles: walking with a comfortable walking speed presented a low fall-risk profile; carrying a one-sided load and walking at a faster speed on the I-beam both presented ...

The validation of gait-stability metrics to assess construction workers' fall risk

Falling from height is the top cause of injuries and fatalities in the construction industry. Understanding the fall risk at different work environments can help to prevent fall accidents on a jobsite. While many previous studies attempted to assess the fall risk on a construction site, most of them are qualitative or subject to cognitive biases. In this context, this paper aims to introduce and validate a quantitative measure that allows researchers to characterize the fall risks of construction workers. In particular, this paper focuses on validating the fall risk predictive power of Maximum Lyapunov exponent (Max LE), which is one of the gait-stability metrics established in clinical settings. The kinematic data were collected using an inertial measurement unit (IMU) sensor attached to the right ankle of the subject performing different tasks. The Max LE for each tasks were then calculated based upon the IMU measurements. The results indicated a significant difference in the Max LE between different tasks, which indicates that Max LE has the potential to evaluate the dynamic stability of construction workers.

EEG Signal-Processing Framework to Obtain High-Quality Brain Waves from an Off-the-Shelf Wearable EEG Device

AbstractInvestigating brain waves collected by an electroencephalogram (EEG) can be useful in understanding human psychosocial conditions such as stress, emotional exhaustion, burnout, and mental f...

Feasibility of Wearable Electromyography (EMG) to Assess Construction Workers’ Muscle Fatigue

Due to the labor-intensive nature of construction tasks, a large number of construction workers frequently suffer from excessive muscle fatigue. Workers’ muscle fatigue can adversely affect their productivity, safety, and well-being. Several attempts have been made to assess workers’ fatigue using subjective methods (e.g., fatigue questionnaire). Despite, the success of subjective methods in assessing workers’ fatigue in a long period, these methods have limited utility on construction sites. For instance, these methods interrupt workers’ ongoing tasks. These methods are also subject to high biases. To address these issues, this study aims to examine the feasibility of a wearable Electromyography (EMG) sensor to measure the electrical impulses produced by workers’ muscles as a means to continuously evaluate workers muscle fatigue without interfering with their ongoing tasks. EMG signals were acquired from eight subjects while lifting a concrete block using their upper limbs (i.e., elbow and shoulder muscles). As the first step, filtering methods (e.g., bandpass filter, rolling filter, and Hampel filter) were applied to reduce EMG signal artifacts. After removing signal artifacts, to examine the potential of EMG in measuring workers’ muscle fatigue, various EMG signal metrics were calculated in time domain (e.g., Signal Mean Absolute Value (MAV) and Root Mean Square (RMS)) and frequency domain (e.g., Median Frequency (MDF) and Mean Frequency (MEF)). Subjects’ perceived muscle exertion (Borg CR-10 scale) was used as a baseline to compare the muscle exertion identified by EMG signals. Results show a significant difference in EMG parameters while subjects were exerting different fatigue levels. Results confirm the feasibility of the wearable EMG to evaluate workers’ muscle fatigue as means for assessing their physical stress on construction sites.

Mobile EEG-Based Workers’ Stress Recognition by Applying Deep Neural Network

A large number of construction workers are struggling with high stress associated with their perilous job sites. Excessive occupational stress can cause serious job difficulties by negatively impacting workers’ productivity, safety, and health. The first step to decrease the adverse outcomes of this work-related stress is to measure workers’ stress and detect the factors causing stress among workers. Various self-assessment instruments (e.g., a stress assessment questionnaire) have been used to assess workers’ perceived stress. However, these methods are compromised by several drawbacks that limit their use in the field. Firstly, these methods interrupt workers ongoing tasks. Secondly, these methods are subject to a high degree of bias, which can lead to inconsistent results. The authors’ earlier work attempted to address the limitations of these subjective methods by applying different machine learning methods (e.g., Supervised Learning algorithms) to identify the pattern of workers’ brain waves that is acquired from a wearable Electroencephalography (EEG) device, while exposed to different stressors. This research thus attempts to improve the stress recognition accuracy of the previous algorithms by developing an EEG-based stress recognition framework by applying two Deep Learning Neural Networks (DNN) structures: a convolutional deep learning neural network (deep CNN) and a Fully Connected Deep Neural Network. Results of the optimum DNN configuration yielded a maximum of 86.62% accuracy using EEG signals in recognizing workers’ stress, which is at least six percent more accurate when compared with previous handcraft feature-based stress recognition methods. Detecting workers’ stress with a high accuracy in the field will lead to enhancing workers’ safety, productivity, and health by early detection and mitigation of stressors at construction sites.