OsteoAI: A Review of Explainable Artificial Intelligence, Osteoporosis Risk Prediction and Preventive Lifestyle Simulation

The reviewed literature shows a growing role for artificial intelligence in osteoporosis classification, fracture-risk prediction, bone-property assessment, and clinical decision support. Machine learning and deep learning provide ways to analyze complex clinical and imaging information, while XAI methods can make individual model outputs easier to interpret. However, important challenges remain, particularly in validation, dataset heterogeneity, reporting quality, generalizability, and integration into clinical practice. OsteoAI builds on these research directions by proposing an explainable osteoporosis risk-assessment framework based on clinical and lifestyle information, supported by personalized preventive guidance and lifestyle simulation. The literature provides a clear rationale for this direction, but it also makes clear that rigorous validation and careful clinical positioning are essential before such a system could be considered for real-world healthcare use.

[1]         G. Gatineau, E. Shevroja, C. Vendrami, E. Gonzalez-Rodriguez, W. D. Leslie, O. Lamy, and D. Hans, “Development and reporting of artificial intelligence in osteoporosis management,” Journal of Bone and Mineral Research, vol. 39, no. 11, pp. 1553–1573, 2024, doi: 10.1093/jbmr/zjae131. [2]         A. Conforti, M. Ruggiero, L. Lucchetti, V. Cipolloni, F. D. Galati, M. Gentile, and A. Lo Gullo, “Artificial Intelligence and Machine Learning in the Diagnosis and Management of Osteoporosis: A Comprehensive Review,” Medicina, vol. 62, no. 1, 27, 2025, doi: 10.3390/medicina62010027. [3]         Y. Wu, J. Chao, M. Bao, and N. Zhang, “Predictive value of machine learning on fracture risk in osteoporosis: a systematic review and meta-analysis,” BMJ Open, vol. 13, no. 12, e071430, 2023, doi: 10.1136/bmjopen-2022-071430. [4]         V. V. Khanna, K. Chadaga, N. Sampathila, R. Chadaga, S. Prabhu, S. K. S., A. S. Jagdale, and D. Bhat, “A decision support system for osteoporosis risk prediction using machine learning and explainable artificial intelligence,” Heliyon, vol. 9, no. 12, e22456, 2023, doi: 10.1016/j.heliyon.2023.e22456. [5]         R. W. Liu, W. Ong, A. Makmur, N. Kumar, X. Z. Low, S. Ge, Y. L. Tan, D. F. K. Ting, J. H. Tan, and J. T. P. D. Hallinan, “Application of Artificial Intelligence Methods on Osteoporosis Classification with Radiographs—A Systematic Review,” Bioengineering, vol. 11, no. 5, 484, 2024, doi: 10.3390/bioengineering11050484. [6]         M. Sadat-Ali, B. A. Alzahrani, T. S. Alqahtani, M. A. Alotaibi, A. M. Alhalafi, A. A. Alsousi, and A. M. Alasiri, “Accuracy of artificial intelligence in prediction of osteoporotic fractures in comparison with dual-energy X-ray absorptiometry and the Fracture Risk Assessment Tool: A systematic review,” World Journal of Orthopedics, vol. 16, no. 4, 103572, 2025, doi: 10.5312/wjo.v16.i4.103572. [7]         Z. Zhang, Y. He, Z. Mo, P. Zhang, Z. Tian, and L. Huang, “Artificial Intelligence for Osteoporosis Diagnosis, Risk Prediction and Therapy: Current Advances, Clinical Challenges, and Future Perspectives,” Clinical Interventions in Aging, vol. 21, 607232, 2026. [8]         S.-T. Fan, M. Lu, J.-L. Dong, Y.-L. Li, L.-N. Hao, R.- C. Dong, and M.-D. Hou, “Application of artificial intelligence in osteoporosis: a review,” Frontiers in Medicine, vol. 12, 1718554, 2025, doi: 10.3389/fmed.2025.1718554.