BMEye: Public Health-Oriented Body Mass Index Monitoring Using Commodity WiFi

Body Mass Index (BMI) is a commonly used measure of body fat based on an individual's height and weight. Public health authorities and organizations monitor BMI at a population level to assess the prevalence and trends of under-weight and obesity within communities. This helps identify at-risk populations, evaluate the effectiveness of interventions, and develop targeted strategies to promote healthy weight and prevent associated health problems. However, there is a lack of BMI monitoring methods that do not require active cooperation or cause privacy concerns to human subjects. This paper introduces a BMI classification approach based on passive WiFi sensing. Our proposed approach is inspired by the observation that different body shapes cause dissimilar impacts on WiFi signal propagation, which can be reflected by Channel State Information (CSI) readings. Leveraging state-of-the-art machine learning algorithms, we developed a BMI sensing system that classifies the BMI of human subjects when they move in a WiFi range. Our evaluation results show that our approach could be an effective tool for public-health-oriented BMI monitoring, yielding a precision rate of 70.86%, a recall rate of 71.21 %, and an F1 score of 73.66%.