This research, presented by Li et. al, examined the application of machine learning models Random Forest (RF) and SHAP analysis on assessing the prevalence of health risks from physical hazards, heat and noise, in the construction industry. Though the concerns associated with heat and noise individually have been studied, the researchers wanted to investigate how co-exposure affects health with white blood cells used as an inflammatory marker of the body’s response to these hazards. In a cross-sectional surveillance study of 10,275 participants from the the Fujian Workplace Occupational Hazards Comprehensive Surveillance Program, the health of workers without baseline diseases from multiple blue collar disciplines was observed between January 1, 2020 and December 31, 2022. Groups that had been exposed to both heat and noise hazards had the most deteriorated health with 38.2% hypertension and 37.8% abnormal ECG. Renal, pulmonary, and liver dysfunction were all consequences of heat exposure but when combined with noise, the odd ratio (OR) of pulmonary dysfunction increased by 0.09, hypertension OR increased by 0.36, and abnormal ECG OR increased by 0.58. White blood cell count was highest in co-exposure groups (~8.32) compared to the heat exposed (~6.76) or noise exposed (6.63) groups. The Random forest model gave accurate guesses as to the prevalence of health deterioration from these hazards. Li et. al acknowledged the lack of geographical diversity in this study and the influence of other detrimental health habits on the data.