A Clinically-Guided Machine Learning Framework for Operational Health Risk Tier Forecasting in Construction Workers Using Wearable Data

The project aims to predict the health risk levels of construction workers using a machine learning framework. By leveraging core physiological indicators such as heart rate, body temperature, and oxygen saturation (SpO2), the system evaluates the workers' health and classifies them into risk categories. The dataset, sourced from real-world anonymized health data, provides input parameters like respiratory rate, oxygen saturation, systolic blood pressure, heart rate, temperature, and consciousness, with the output being the predicted health risk level. Existing algorithms, such as K-means clustering, logistic regression, random forest, and XGBoost, are explored, while proposed algorithms include decision trees, CatBoost, SVM, and TabPFNClassifier. The backend of the system is developed using Python with Flask, and the frontend is built using HTML, CSS, and JavaScript, with MySQL being used for database management. The primary goal is to provide a clinical tool for accurately forecasting health risks in workers based on wearable data, thus enabling better management of health and safety in construction environments.

Keywords:

Health Risk Prediction , Construction Workers , Machine Learning ,  Decision Tree, CatBoost , SVM , TabPFNClassifier , K-means Clustering , Random Forest , XGBoost , Flask