Application of artificial intelligence in osteoporosis: a review

Shuting Fan, Minfeng Lü, Jianming Dong, Yilin Li, Lina Hao, Rong Dong, Min HouFrontiers in Medicine 2025 · DOI

Abstract

Osteoporosis is a systemic skeletal disease defined by decreased bone mass, deteriorated bone microarchitecture, and increased fracture susceptibility, all of which substantially increase the risks of functional disability and overall mortality. Given its complex pathophysiology, chronic progression, and frequently asymptomatic clinical presentation, the diagnosis and treatment of osteoporosis present multiple challenges. Artificial intelligence (AI) technologies integrate radiomics and multi-omics data to develop intelligent frameworks that map disease progression, identify molecular biomarkers, and stratify individual risk, offering new strategies for molecular-level support in precision diagnosis, preventive monitoring, and personalized intervention. Notably, AI-driven models for drug target prediction and therapeutic response evaluation have also been developed, offering mechanism-based strategies for novel drug discovery. This study outlines how AI advances osteoporosis management by applying technological breakthroughs in foundational and clinical research across key areas, including optimizing preventive monitoring, developing precision diagnostics, supporting personalized treatment decisions, and enabling molecular-targeted drug discovery, thereby offering a strong theoretical basis and practical pathways within precision and personalized medicine.

Introduction: Addressing the "Silent" Epidemic

Osteoporosis (OP) remains a significant global health challenge, often referred to as a "silent" disease because it frequently remains asymptomatic until a fragility fracture occurs. Traditional diagnostic methods, primarily Dual-energy X-ray Absorptiometry (DXA), are the current gold standard but have notable limitations in sensitivity and fail to capture the multi-faceted biological drivers of bone loss. This review explores the transformative role of Artificial Intelligence (AI) in shifting osteoporosis management toward a more proactive, personalized, and precise paradigm. By bridging the gap between molecular research and clinical practice, AI offers the potential to predict fractures earlier and tailor treatments to individual patient profiles.

AI Methodologies and Multi-Omics Integration

The application of AI in osteoporosis spans a broad spectrum of methodologies, from traditional machine learning to advanced deep learning architectures.

  • Machine Learning (ML): Algorithms such as Support Vector Machines (SVM), Random Forests (RF), and Gradient Boosting Machines (GBM) are instrumental in processing high-dimensional "omics" data. These models excel at feature engineering, allowing researchers to identify critical molecular biomarkers from complex datasets.
  • Deep Learning (DL): Convolutional Neural Networks (CNNs) have revolutionized automated image analysis, while Graph Neural Networks (GNNs) are increasingly used to map intricate drug-target interactions.
  • Integrative Frameworks: The most promising advancements involve multimodal AI systems that fuse radiomics (imaging data) with genomics, proteomics, and clinical profiles to create a comprehensive view of patient health.
Workflow of AI-driven multi-omics analysis for identifying biomarkers and therapeutic targets.

Key Insights: Beyond Bone Density

AI-driven research has moved beyond simple density measurements to uncover the underlying biological networks of osteoporosis.

  • Molecular Mechanisms: AI has identified core links between bone loss and immune system imbalances, as well as programmed cell death pathways like apoptosis and ferroptosis.
  • Cross-Disease Connections: Sophisticated algorithms have revealed molecular overlaps between OP and conditions such as Inflammatory Bowel Disease (IBD) and Hepatitis B, as well as the impact of environmental factors like PFAS exposure.
  • Opportunistic Screening: AI models can analyze routine medical images—such as dental X-rays or chest CTs taken for unrelated reasons—to identify patients at risk for OP, enabling early intervention.
  • Novel Therapeutics: Deep learning systems like Deep Transformers have been used to screen traditional herbal compounds, identifying dozens of active ingredients for potential drug development.

Clinical Impact and Predictive Performance

The clinical utility of AI in osteoporosis is supported by impressive performance metrics across various diagnostic and prognostic tasks. Diagnostic accuracy has seen significant gains; for instance, hybrid deep learning frameworks have achieved an Area Under the Curve (AUC) of 0.97 and 96% accuracy in predicting bone mineral density from biplanar X-rays. Furthermore, models for classifying vertebral fractures have reached an AUC of 0.948, matching or exceeding the performance of expert radiologists. AI-assisted systems have also been shown to improve clinicians' diagnostic accuracy by up to 18.5%, demonstrating their value as a collaborative tool in the clinic.

Deep learning architectures for multi-modal data integration in clinical osteoporosis management.

Challenges and Future Directions

Despite these advancements, several barriers to widespread clinical adoption remain. "Model drift"—where performance degrades when applied to new, real-world datasets—and the "black box" nature of complex neural networks pose challenges for clinical trust and interpretability. Furthermore, the integration of sensitive multi-omics data necessitates robust privacy frameworks.

Future research is expected to focus on federated learning, which allows models to be trained across multiple institutions without sharing raw patient data, and a continued emphasis on moving these powerful tools from the laboratory into routine clinical workflows. By addressing these hurdles, AI can truly realize its potential in delivering personalized medicine for osteoporosis patients.

Summary updated

Abstract

Osteoporosis is a systemic skeletal disease defined by decreased bone mass, deteriorated bone microarchitecture, and increased fracture susceptibility, all of which substantially increase the risks of functional disability and overall mortality. Given its complex pathophysiology, chronic progression, and frequently asymptomatic clinical presentation, the diagnosis and treatment of osteoporosis present multiple challenges. Artificial intelligence (AI) technologies integrate radiomics and multi-omics data to develop intelligent frameworks that map disease progression, identify molecular biomarkers, and stratify individual risk, offering new strategies for molecular-level support in precision diagnosis, preventive monitoring, and personalized intervention. Notably, AI-driven models for drug target prediction and therapeutic response evaluation have also been developed, offering mechanism-based strategies for novel drug discovery. This study outlines how AI advances osteoporosis management by applying technological breakthroughs in foundational and clinical research across key areas, including optimizing preventive monitoring, developing precision diagnostics, supporting personalized treatment decisions, and enabling molecular-targeted drug discovery, thereby offering a strong theoretical basis and practical pathways within precision and personalized medicine.
BibTeX
@inproceedings{fan2025application,
  author = {Shuting Fan and Minfeng Lü and Jianming Dong and Yilin Li and Lina Hao and Rong Dong and Min Hou},
  title = {{Application of artificial intelligence in osteoporosis: a review}},
  booktitle = {Frontiers in Medicine},
  year = {2025},
  doi = {10.3389/fmed.2025.1718554},
}
Figure