A Bibliometric Analysis of Neuroscience Tools Use in Construction Health and Safety Management.

Despite longstanding traditional construction health and safety management (CHSM) methods, the construction industry continues to face persistent challenges in this field. Neuroscience tools offer potential advantages in addressing these safety and health issues by providing objective data to indicate subjects' cognition and behavior. The application of neuroscience tools in the CHSM has received much attention in the construction research community, but comprehensive statistics on the application of neuroscience tools to CHSM is lacking to provide insights for the later scholars. Therefore, this study applied bibliometric analysis to examine the current state of neuroscience tools use in CHSM. The development phases; the most productive journals, regions, and institutions; influential scholars and articles; author collaboration; reference co-citation; and application domains of the tools were identified. It revealed four application domains: monitoring the safety status of construction workers, enhancing the construction hazard recognition ability, reducing work-related musculoskeletal disorders of construction workers, and integrating neuroscience tools with artificial intelligence techniques in enhancing occupational safety and health, where magnetoencephalography (EMG), electroencephalography (EEG), eye-tracking, and electrodermal activity (EDA) are four predominant neuroscience tools. It also shows a growing interest in integrating the neuroscience tools with artificial intelligence techniques to address the safety and health issues. In addition, future studies are suggested to facilitate the applications of these tools in construction workplaces by narrowing the gaps between experimental settings and real situations, enhancing the quality of data collected by neuroscience tools and performance of data processing algorithms, and overcoming user resistance in tools adoption.

Monitoring Inattention in Construction Workers Caused by Physical Fatigue Using Electrocardiograph (ECG) and Galvanic Skin Response (GSR) Sensors.

Physical fatigue is frequent for heavy manual laborers like construction workers, but it causes distraction and may lead to safety incidents. The purpose of this study is to develop predictive models for monitoring construction workers' inattention caused by physical fatigue utilizing electrocardiograph (ECG) and galvanic skin response (GSR) sensors. Thirty participants were invited to complete an attention-demanding task under non-fatigued and physically fatigued conditions. Supervised learning algorithms were utilized to develop models predicting their attentional states, with heart rate variability (HRV) features derived from ECG signals and skin electric activity features derived from GSR signals as data inputs. The results demonstrate that using HRV features alone could obtain a prediction accuracy of 88.33%, and using GSR features alone could achieve an accuracy of 76.67%, both through the KNN algorithm. The accuracy increased to 96.67% through the SVM algorithm when combining HRV and GSR features. The findings indicate that ECG sensors used alone or in combination with GSR sensors can be applied to monitor construction workers' inattention on job sites. The findings would provide an approach for detecting distracted workers at job sites. Additionally, it might reveal the relationships between workers' physiological features and attention.