Artificial Intelligence for Opportunistic Screening for Osteoporosis and Spine Fractures Using Computed Tomography: A
Farid Ghareh Mohammadi1,2, Ronnie Sebro1,2,3
1Department of Radiology.
Journal of Computer Assisted Tomography
|June 29, 2026
Summary
Artificial intelligence (AI) algorithms show excellent performance in screening for osteoporosis and fractures using computed tomography (CT) scans. This opportunistic approach can improve early detection and prevention of bone conditions.
Area of Science:
- Radiology and Medical Imaging
- Artificial Intelligence in Healthcare
- Bone Health and Osteoporosis
Background:
- Osteoporosis is a significant global health issue, often detected late after fractures occur.
- Dual-energy X-ray absorptiometry (DXA) is underutilized in clinical practice.
- Computed tomography (CT) scans offer a potential for opportunistic bone quality assessment without additional radiation.
Purpose of the Study:
- To systematically review and meta-analyze the diagnostic performance of AI algorithms for osteoporosis screening using CT.
- To evaluate AI's performance in fracture screening using CT.
- To identify the most studied body parts and predominant AI algorithms (ML/DL).
Main Methods:
- Systematic review and meta-analysis following PRISMA guidelines, searching major databases (PubMed, Web of Science, Scopus, Google Scholar) from 2011-2025.
- Inclusion of studies using DXA as the reference standard and reporting diagnostic performance metrics.
- Random-effects meta-analysis to pool sensitivity, specificity, and AUC, with subgroup analyses by algorithm type and study size.
Main Results:
- 31 studies were included, focusing on osteoporosis (83.87%) and fracture (16.13%) screening.
- The abdomen/lumbar spine (L-spine) was the most frequently evaluated body part for both osteoporosis and fracture screening.
- Pooled AUC for osteoporosis screening was 0.931 and for fracture screening was 0.863. Machine learning (ML), particularly Support Vector Machines (SVM), predominated over deep learning (DL).
Conclusions:
- AI algorithms applied to abdomen/L-spine CT scans demonstrate excellent diagnostic performance for opportunistic osteoporosis and fracture screening.
- Integration of these AI tools into routine CT interpretation can significantly improve early detection and prevention strategies for bone diseases.
- The findings support the potential of AI in leveraging existing CT scans for enhanced bone health assessment.
