OSTEO-AI: A Systematic Review and Meta-Analysis of Artificial Intelligence Models for Osteoarthritis and Osteoporosis Detection and Prognosis
Bibliographic record
Abstract
Introduction: Osteoarthritis (OA) and osteoporosis are leading degenerative bone diseases that diminish quality of life and impose significant socioeconomic costs. Traditional diagnostic approaches, including imaging and bone density assessments, often fail to detect disease in its early stages, delaying critical interventions. Emerging artificial intelligence (AI) techniques, particularly those employing machine learning (ML) and deep learning (DL), offer promising avenues for early detection and more accurate prognostication. Methods: We conducted a systematic review of AI models developed between 2018 and 2024, assessing their performance in diagnosing and predicting the progression of OA and osteoporosis. Studies utilizing supervised or unsupervised methods applied to imaging modalities (e.g., X-ray, MRI, DXA) or clinical data were included. We evaluated model accuracy, reliability, clinical applicability, and generalizability. Quality and risk of bias were assessed using a modified CLAIM framework, ensuring alignment with transparency, validity, and clinical integration standards. Results: Of 2,300 identified articles, 33 studies met the inclusion criteria. Top-performing models for OA reached up to 97% accuracy, with one study achieving an AUC of 0.93 for MRI-based progression prediction. For osteoporosis, the strongest models attained a C-index of 0.90 using DXA imaging, indicating robust fracture risk prediction. Nevertheless, many studies relied on geographically or demographically homogeneous datasets, limiting broader applicability. Only 15% included external validation, and a substantial proportion lacked interpretability features essential for clinical adoption. Discussion: AI-driven models outperformed conventional diagnostic tools in accuracy and early disease detection. However, the limited dataset diversity, infrequent external validation, and insufficient model interpretability pose barriers to clinical integration. The reliance on male-dominant datasets for osteoporosis and geographically narrow cohorts for OA underscores the need for broader data representation. Standardizing evaluation metrics and improving explainability will enhance cross-study comparisons and support adoption in practice. Conclusion: AI holds transformative potential for improving OA and osteoporosis diagnostics, facilitating earlier interventions, and informing personalized patient management. Future work should prioritize diverse, well-validated datasets; transparent, clinician-friendly interfaces; and standardized performance metrics. Addressing these challenges will enable AI to evolve from a promising innovation into a cornerstone of global musculoskeletal healthcare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.021 | 0.064 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.014 | 0.031 |
| Bibliometrics | 0.010 | 0.010 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".