Applications of Artificial Intelligence in Opportunistic Screening and Diagnostic Imaging of Osteoporotic Fractures: A Systematic Review

Abstract

Background: Osteoporosis remains a "silent epidemic," with 80% of high-risk patients untreated. This systematic review assesses the clinical utility of Artificial Intelligence (AI) in opportunistic screening. It examines how routine CT and MRI scans identify vertebral fractures and low bone mineral density (BMD) without an extra dose of radiation.

Methods: Using PRISMA guidelines, we systematically searched major databases and identified 30 studies focused on Deep Learning and Convolutional Neural Networks (CNNs) in clinical procedures.

Results: AI models achieved Area Under the Curve (AUC) between 0.81 and 0.99. About 50% of studies used routine chest and abdominal CT scans, in which AI-driven 3D volumetric analysis outperformed traditional 2D DXA densitometry (AUC 0.82 vs. 0.72). The algorithms reduced diagnostic errors from aortic calcifications and degenerative changes and enabled the prediction of 5- to 10-year fracture risk.

Conclusions: Opportunistic screening in connection with AI-Based programs closes the diagnostic gap in bone and skeletal system screening and diagnostics. Integrating automated vertebral checks into daily imaging workflows supports early diagnosis. This proactive approach improves long-term patient health and reduces the financial burden on the healthcare system.