Novel Approach To Predict Core Temperature For Workers With High High-level Personal Protective Equipment

PURPOSE: This study aimed to create a novel, non-invasive system to predict core temperature (Tc) among occupational workers who wear high-level personal protective equipment and perform physical work. The direct measurement of Tc is typically accomplished through invasive techniques (rectal, esophageal, or intestinal). Other predictive methods have been found but usually involve complex, non-deployable, or inaccurate systems. Here, we hypothesized that a supervised machine learning model using regression and variables collected from a single piece of commercially available physiological monitoring equipment could effectively predict Tc. METHODS: Forty-five adult participants (43 male, 2 female) exercised on a treadmill in two conditions while wearing firefighter personal protective equipment. The first condition (14 subjects) was a graded exercise protocol in a normothermic environment (25-26 °C, 45% humidity). The second condition (31 subjects) was a steady-state exercise protocol in a hyperthermic condition (35-36 °C, 50% humidity). The variables collected using a physiological monitoring system were skin temperature (Tsk), heart rate (HR), time, respiratory rate (RR), and rate of skin temperature acquisition per minute (Tsk/min). A total of 3629 data points were used in the supervised machine learning regression analysis. 80% of the data was used to train the predictive model. The model was validated using the remaining 20% of the data. The model's performance was measured using mean absolute error (MAE) and root mean squared error (RMSE). RESULTS: The predictive model created in this study had an R2 of 0.87, and the model's performance resulted in an MAE of 0.18 °C and an RMSE of 0.23 °C. CONCLUSION: This Tc prediction method creates a practical, non-invasive model to predict Tc with minimal resources. This method can continuously monitor workers who wear high-level personal protective equipment and perform physical duties. This monitoring can prevent heat-related morbidity and mortality by preventing workers from reaching detrimental heat stress-related states.