Test-retest reliability, validity, and responsiveness of a textile-based wearable sensor for real-time assessment of physical fatigue in construction bar-benders

Highlights

• Physiological parameters were evaluated to assess real-time physical fatigue.

• The textile-based wearable sensors are a reliable method to assess physical fatigue.

• Physiological parameters were correlated with subjective fatigue in bar-benders.

• Physiological parameters and blood lactate levels were unrelated in bar-benders.

Abstract

While recent studies have shown that wearable sensing technology has the potential to facilitate the evaluation of physical fatigue, the reliability and validity of such measurements during construction tasks have not been reported. Thus, the primary objective of the current study is to establish absolute and relative reliability of textile-based wearable sensors to monitor physical fatigue during bar bending and fixing construction tasks. The secondary objective is to establish correlations between physiological parameters and subjective fatigue scores or blood lactate levels in order to demonstrate the convergent validity. Physiological parameters such as heart rate, breathing rate, and skin temperature were evaluated using textile-based wearable sensors. The test-retest reliability (intra-class correlation coefficient - ICC) values of the measured resting and working heart rate (ICC = 0.73 and 0.85), breathing rate (ICC = 0.78 and 0.82), and skin temperature (ICC = 0.68 and 0.77) were moderate to good and good, respectively. There were moderate to excellent correlations (r-values ranging from 0.414 to 0.940) between physiological parameters and subjective fatigue scores, although there were no correlations between any physiological parameters and blood lactate levels. Both laboratory and field data substantiated that the wearable sensing system has the potential to be a reliable noninvasive device to monitor physical fatigue (especially among workers at risk of sustaining fatigue-related injury due to advanced age, poor health, or job nature. However, because the current study validated the system exclusively in bar benders, additional research is necessary to confirm the findings in other construction workers.

Introduction

Fatigue is defined as the effect of continued work, weariness, or exhaustion of physical or mental strength that may result in a temporary loss of ability to work [1]. Around 40% of construction workers in the USA have reported experiencing extreme fatigue, which could have a negative impact on worker safety, general health, and overall productivity [2]. Long working hours, hot and humid work environments, and heavy workloads have been shown to exacerbate the detrimental effects of fatigue in construction workers [1,3,4]. Additionally, excessive fatigue may increase the risk of work-related musculoskeletal disorders and absenteeism in construction workers [5,6]. Thus, assessing physical fatigue in construction workers is critical as the first step toward minimizing their risk of physical fatigue.

Wearable sensors have been widely used to monitor physiological parameters in many industries such as health [[7], [8], [9], [10], [11]], sports [12,13], mining [14,15], and construction [[16], [17], [18], [19], [20]]. Within each industry, several commercial wearable sensors are currently available [8]. Recent advancements in wireless technology, Internet of Things, and miniature sensors have made possible a new generation of monitoring systems that can record physiological data from individuals without interfering with their daily activities in uncontrolled environments [21]. The application of wearable sensors in the construction industry is in its infancy in comparison to other industries. Recently, some studies have employed wearable sensors to assess physical exertion and fatigue via monitoring physiological parameters including heart rate (HR), breathing rate, and skin temperature in construction workers [[17], [18], [19],22,23]. For instance, Yi et al. [22] and Aryal et al. [17] assessed physical fatigue in construction workers using HR and skin temperature metrics. The classification accuracy was 9% higher when only features extracted from average skin temperature data were used, compared to when only heart rate data were used, and combining data from both sensors resulted in the highest accuracy of 82% [17]. Additionally, Anwer et al. [18] also found positive correlations between physiological parameters and subjective fatigue scores.

The development of new wearable technologies and latest progress in physiology have allowed real-time objective monitoring of physical exertion and fatigue during construction tasks. While some wearable devices can capture only one or two parameters, other devices can capture multiple parameters simultaneously [24]. For instance, the Equivital Lifemonitor (EQ02) is a wearable ambulatory device used to measure numerous physiological parameters including HR, skin temperature, and breathing rate via a chest-worn textile with embedded sensors [25]. Previous studies have evaluated the accuracy and validity of the EQ02 wearable sensor system for monitoring HR, skin temperature, and breathing rate in healthy young adults [[25], [26], [27]]. However, only one study tested the reliability of this device in monitoring physiological parameters at rest and during activities of low to moderate intensity [26]. Although the EQ02 wearable sensor system (EQ02 system hereafter) has been used extensively to examine heat strain, physical exertion, and fatigue through monitoring of physiological parameters in construction workers [[17], [18], [19],[28], [29], [30], [31]], no study has examined the reliability and validity of the EQ02 system in evaluating physiological parameters for the real-time assessment of physical fatigue during actual construction tasks. Additionally, Akintola et al. [25] found that the EQ02 system had high quantity of movement artifacts, which may affect the usage of this device during actual construction tasks. Importantly, previous studies did not examine the associations between changes in physiological parameters and objective fatigue biomarkers such as blood lactate levels during construction activities [[17], [18], [19]]. Since blood lactate levels could predict fatigue during high intensity physical exercise [32], it is important to establish the reliability, validity, and responsiveness of the EQ02 system in monitoring physiological parameters for real-time fatigue assessment before deploying it for use in on-site monitoring of construction activities. To bridge these gaps, the objectives of this study are to evaluate the test-retest reliability, validity, and responsiveness of the EQ02 system for assessing physiological parameters during a bar bending and fixing construction tasks. The major contribution of the current study is the evaluation of the performance of a chest-worn textile with embedded sensors (EQ02 system) for real-time physical fatigue assessment in construction workers.