Predicting Osteoporosis Prevalence Based on Ambient-Air Pollution Using Artificial Intelligence

Ambient-air pollution is a recognized risk factor for numerous health problems, including osteoporosis. Previous studies demonstrated that air pollution, particularly that caused by particulate matter, can be a modifiable risk factor for osteoporosis. Therefore, this study proposes an artificial-intelligence-based model to predict the prevalence of osteoporosis caused by air-pollution exposure in South Korea. We conducted a retrospective cohort study using publicly available data from 2015 to 2023. Air-quality data were obtained from the Korean Statistical Information Service, and the number of osteoporosis was retrieved from the Health Insurance Review and Assessment Service database. A deep neural network (DNN) model was developed to predict the prevalence of osteoporosis. The model demonstrated promising prediction accuracies ranging from 85.31% to 86.78%, with a mean absolute percentage error of 13.95%. These findings suggest the potential application of DNN models for predicting the prevalence of osteoporosis based on ambient-air pollution. The present study serves as a valuable foundation for the further investigation and development of predictive models for osteoporosis.