I-Corps: Translation Potential of an Artificial Intelligence (AI)-Based Platform for Early Osteoporosis Risk Prediction

Award ID: 2630770

This I-Corps project is based on the development of an artificial intelligence (AI)-based software platform to predict osteoporosis risk. Osteoporosis is a major health challenge affecting millions of older adults. However, osteoporosis screening is not routinely integrated into clinical workflows, so the disease typically remains undetected until a fracture occurs. These fractures create significant healthcare costs and negatively impact patient quality of life. This technology evaluates osteoporosis risk with high patient-level accuracy within a few seconds to minutes, reducing physician evaluation time and enabling earlier identification of patients at high risk of osteoporotic fractures. In addition, the platform is designed to integrate with existing imaging and health record systems at hospitals, imaging centers, and healthcare networks. This technology may allow both clinicians and individuals to assess their osteoporosis risk, receive preventive recommendations, support early detection, and help reduce healthcare costs associated with fractures and hospitalizations. Timely risk assessment may support preventive interventions that can improve patient outcomes while reducing the clinical and economic burden of fractures and associated healthcare costs. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of an artificial intelligence (AI) platform for osteoporosis risk screening. Prior work has focused primarily on fracture prediction, single anatomical sites, or limited patient populations. This technology is an AI-based framework that integrates machine learning (ML) and deep learning (DL) models to analyze a patients clinical data from electronic health records and computed tomography (CT) imaging data and evaluate osteoporosis risk. Previous research in AI-driven clinical risk prediction using clinical variables and CT-derived radiomic features to train and evaluate ML and DL models demonstrated promising model performance, establishing the technical feasibility of integrating clinical, laboratory, and imaging-derived data for accurate risk prediction. By enabling automated, low-cost screening from routine CT scans, the platform may identify high-risk patients earlier, facilitate timely intervention, reduce osteoporotic fractures, and lower healthcare costs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman