{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T18:32:52Z","timestamp":1792521172078,"version":"4.5.0"},"reference-count":44,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2026,8,6]],"date-time":"2026-08-06T00:00:00Z","timestamp":1785974400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100005375","name":"Latvian Council of Science","doi-asserted-by":"crossref","award":["LV-UA\/2025\/4"],"award-info":[{"award-number":["LV-UA\/2025\/4"]}],"id":[{"id":"10.13039\/501100005375","id-type":"DOI","asserted-by":"crossref"}]},{"award":["LV-UA\/2025\/4"],"award-info":[{"award-number":["LV-UA\/2025\/4"]}],"id":[{"id":"https:\/\/ror.org\/02ar66p97","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100007684","name":"Ministry of Education and Science of Ukraine","doi-asserted-by":"crossref","award":["0125U003179"],"award-info":[{"award-number":["0125U003179"]}],"id":[{"id":"10.13039\/501100007684","id-type":"DOI","asserted-by":"crossref"}]},{"award":["0125U003179"],"award-info":[{"award-number":["0125U003179"]}],"id":[{"id":"https:\/\/ror.org\/03bra4w96","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Medicina"],"abstract":"<jats:p>Background and Objectives: Osteoporotic vertebral compression fractures affect approximately one in four postmenopausal women and carry substantial morbidity, yet established clinical tools such as dual-energy X-ray absorptiometry (DXA) provide only patient-level risk and do not identify which specific vertebra is most likely to fail. Computed tomography (CT) acquired for unrelated indications is the most widely available three-dimensional substrate for opportunistic screening, but published machine learning models for vertebral fracture risk almost universally operate at the patient level. The present study aimed to develop and rigorously validate a per-vertebra prediction pipeline applicable to both routine clinical lumbar-spine CT and opportunistic abdominal CT, both acquired for indications unrelated to osteoporosis screening. Materials and Methods: Two independent retrospective cohorts were assembled from a single academic centre: a routine clinical lumbar-spine CT cohort of 106 patients yielding 478 evaluable vertebrae, and a routine abdominal CT cohort of 126 patients yielding 589 evaluable vertebrae. Vertebral bodies were segmented automatically with TotalSegmentator v2 and the trabecular core isolated by morphological erosion. A panel of 505 quantitative imaging biomarkers compliant with Image Biomarker Standardisation Initiative recommendations was extracted, covering trabecular density, vertebral morphometry, classical texture, trabecular network architecture, sub-endplate vulnerability, low-density topology, radial heterogeneity and adjacent muscle quality. Within-patient feature engineering expanded the input pool to 1293 contextual descriptors. Three model families were evaluated under fully nested leave-one-patient-out cross-validation: ElasticNet logistic regression, a softmax-ranking approximation of conditional logistic regression, and a Two-Stage model combining a patient-level fragility score with a within-patient vertebral outlier score. Patient-level bootstrap resampling (2000 iterations) was used to obtain 95% confidence intervals. Results: On routine clinical lumbar-spine CT the Two-Stage model achieved a per-vertebra AUC of 0.750 (95% CI 0.704 to 0.795), an F1 of 0.549, a within-patient concordance index of 0.693, an expected calibration error of 0.044, and Hit@3 of 0.934. It was the only model evaluated that returned calibrated probabilities; the softmax-ranking and ElasticNet baselines gave expected calibration errors of 0.232 and 0.218 respectively. On opportunistic abdominal CT, the softmax-ranking model gave AUC 0.672 (95% CI 0.615 to 0.727). Selected biomarkers were dominated by regional trabecular density and trabecular network architecture; a stable core of lumbar features entered the model in 100% of cross-validation folds, indicating high reproducibility. The closest prior per-vertebra CT-based predictor in primary, non-surgical patients (Muehlematter and colleagues, 58-patient cohort) reported a per-vertebra AUC of 0.64, which is one of several reference points for the present results. Ten methodological variants and sensitivity analyses, including rank fusion, internal tissue normalisation and additional biomechanical features, did not provide statistically significant gains, indicating that the binding constraint at this sample size is data volume rather than methodology. Conclusions: A two-stage decomposition that separates systemic skeletal fragility from within-patient vertebral outlier status produces well-calibrated per-vertebra fracture-risk estimates from routine clinical lumbar spine CT and was the only model evaluated to do so, which is what permits a per-vertebra output to be reported as an absolute risk rather than as an ordering alone; a within-patient ranking model is preferable for opportunistic abdominal CT. The discrimination advantage of the decomposition over that baseline is numerical and consistent but not statistically established at this sample size, and the work is presented as a transparent and reproducible single-centre benchmark for the still under-developed per-vertebra prediction task. Its clearest near-term value is opportunistic, namely flagging elevated per-vertebra fracture risk on CTs already acquired for unrelated indications without additional radiation, cost or a dedicated densitometric study. External multi-centre validation is the necessary next step.<\/jats:p>","DOI":"10.3390\/medicina62081518","type":"journal-article","created":{"date-parts":[[2026,8,6]],"date-time":"2026-08-06T13:03:27Z","timestamp":1786021407000},"page":"1518","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Per-Vertebra Prediction of Future Osteoporotic Fractures from Routine Computed Tomography Using a Two-Stage Machine Learning Framework"],"prefix":"10.3390","volume":"62","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8771-8060","authenticated-orcid":false,"given":"Kirill","family":"Riazanovskiy","sequence":"first","affiliation":[{"name":"Department of Artificial Intelligence, Institute for Applied System Analysis, National Technical University of Ukraine \u201cIgor Sikorsky Kyiv Polytechnic Institute\u201d (KPI), 03056 Kyiv, Ukraine"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7428-8592","authenticated-orcid":false,"given":"D\u0101vids","family":"Orlovs","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine, R\u012bgas Stradi\u0146a Universit\u0101te (RSU), LV-1007 R\u012bga, Latvia"},{"name":"Institute of Diagnostic Radiology, Pauls Stradi\u0146\u0161 Clinical University Hospital (PSKUS), LV-1002 R\u012bga, Latvia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-8637-8934","authenticated-orcid":false,"given":"Jekaterina","family":"Stepanova","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine, R\u012bgas Stradi\u0146a Universit\u0101te (RSU), LV-1007 R\u012bga, Latvia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3297-9060","authenticated-orcid":false,"given":"Victor","family":"Sineglazov","sequence":"additional","affiliation":[{"name":"Department of Artificial Intelligence, Institute for Applied System Analysis, National Technical University of Ukraine \u201cIgor Sikorsky Kyiv Polytechnic Institute\u201d (KPI), 03056 Kyiv, Ukraine"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8865-5949","authenticated-orcid":false,"given":"Ardis","family":"Platkajis","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine, R\u012bgas Stradi\u0146a Universit\u0101te (RSU), LV-1007 R\u012bga, Latvia"},{"name":"Institute of Diagnostic Radiology, Pauls Stradi\u0146\u0161 Clinical University Hospital (PSKUS), LV-1002 R\u012bga, Latvia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3006-7460","authenticated-orcid":false,"given":"Olena","family":"Chumachenko","sequence":"additional","affiliation":[{"name":"Department of Artificial Intelligence, Institute for Applied System Analysis, National Technical University of Ukraine \u201cIgor Sikorsky Kyiv Polytechnic Institute\u201d (KPI), 03056 Kyiv, Ukraine"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,6]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"2359","DOI":"10.1007\/s00198-014-2794-2","article-title":"Clinician\u2019s guide to prevention and treatment of osteoporosis","volume":"25","author":"Cosman","year":"2014","journal-title":"Osteoporos. Int."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1726","DOI":"10.1007\/s00198-006-0172-4","article-title":"An estimate of the worldwide prevalence and disability associated with osteoporotic fractures","volume":"17","author":"Johnell","year":"2006","journal-title":"Osteoporos. Int."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1531","DOI":"10.1007\/s00198-017-3909-3","article-title":"Worldwide prevalence and incidence of osteoporotic vertebral fractures","volume":"28","author":"Ballane","year":"2017","journal-title":"Osteoporos. Int."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"949","DOI":"10.2214\/ajr.183.4.1830949","article-title":"Diagnosis of osteoporotic vertebral fractures: Importance of recognition and description by radiologists","volume":"183","author":"Lenchik","year":"2004","journal-title":"Am. J. Roentgenol."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Buckens, C.F., de Jong, P.A., Mol, C., Bakker, E., Stallman, H.P., Mali, W.P., van der Graaf, Y., and Verkooijen, H.M. (2013). Intra- and interobserver reliability and agreement of semiquantitative vertebral fracture assessment on chest computed tomography. PLoS ONE, 8.","DOI":"10.1371\/journal.pone.0071204"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1007\/s11926-020-00892-w","article-title":"Update on imaging-based measurement of bone mineral density and quality","volume":"22","author":"Link","year":"2020","journal-title":"Curr. Rheumatol. Rep."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1016\/j.jocd.2015.08.004","article-title":"Epidemiology of vertebral fractures","volume":"19","author":"Schousboe","year":"2016","journal-title":"J. Clin. Densitom."},{"key":"ref_8","first-page":"CD006349","article-title":"Percutaneous vertebroplasty for osteoporotic vertebral compression fracture","volume":"4","author":"Buchbinder","year":"2018","journal-title":"Cochrane Database Syst. Rev."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"588","DOI":"10.7326\/0003-4819-158-8-201304160-00003","article-title":"Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications","volume":"158","author":"Pickhardt","year":"2013","journal-title":"Ann. Intern. Med."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"860","DOI":"10.1002\/jbmr.3383","article-title":"Future osteoporotic fracture risk related to lumbar vertebral trabecular attenuation at routine body CT","volume":"33","author":"Lee","year":"2018","journal-title":"J. Bone Miner. Res."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"582","DOI":"10.2214\/AJR.20.22874","article-title":"Value-added opportunistic CT: Insights into osteoporosis and sarcopenia","volume":"215","author":"Boutin","year":"2020","journal-title":"Am. J. Roentgenol."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"64","DOI":"10.1148\/radiol.2020200466","article-title":"Automated abdominal CT imaging biomarkers for opportunistic prediction of future major osteoporotic fractures in asymptomatic adults","volume":"297","author":"Pickhardt","year":"2020","journal-title":"Radiology"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Bodden, J., Sollmann, N., El Husseini, M., Sekuboyina, A., L\u00f6ffler, M.T., Zimmer, C., Kirschke, J.S., and Baum, T. (2023). Incidental vertebral fracture prediction using automatic spine segmentation and volumetric BMD extraction from routine clinical CT. Front. Endocrinol., 14.","DOI":"10.3389\/fendo.2023.1207949"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Goller, S.S., Reidler, P., Rudolph, J., Stahl, R., Wulfff, A., and Gersing, A.S. (2023). Automated opportunistic trabecular volumetric bone mineral density extraction outperforms manual measurements for the prediction of vertebral fractures in routine CT. Diagnostics, 13.","DOI":"10.3390\/diagnostics13122119"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: A self-configuring method for deep-learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat. Methods"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"e230024","DOI":"10.1148\/ryai.230024","article-title":"TotalSegmentator: Robust segmentation of 104 anatomic structures in CT images","volume":"5","author":"Wasserthal","year":"2023","journal-title":"Radiol. Artif. Intell."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"323","DOI":"10.1007\/s00198-018-4716-1","article-title":"Prediction of incident vertebral fracture using CT-based finite element analysis","volume":"30","author":"Allaire","year":"2019","journal-title":"Osteoporos. Int."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e48535","DOI":"10.2196\/48535","article-title":"A CT-based fracture prediction model with images of vertebral bones and muscles by employing deep learning","volume":"26","author":"Kong","year":"2024","journal-title":"J. Med. Internet Res."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Kong, S.H., Choi, S., Cho, W., Park, S.B., Park, S.S., Choo, J., Kim, J.H., Kim, S.W., and Shin, C.S. (2025). Enhancing vertebral fracture prediction using multitask deep learning of CT imaging of bone and muscle. Eur. Radiol., online ahead of print.","DOI":"10.1007\/s00330-025-12049-3"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Dieckmeyer, M., L\u00f6ffler, M.T., El Husseini, M., Sekuboyina, A., Menze, B., Sollmann, N., Wostrack, M., Zimmer, C., Baum, T., and Kirschke, J.S. (2022). Level-specific volumetric bone mineral density threshold values for prediction of incident vertebral fractures using opportunistic quantitative CT. Front. Endocrinol., 13.","DOI":"10.3389\/fendo.2022.882163"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Hummel, J., Ferraro, F., Engelke, K., and Chaudry, O. (2025). Trabecular texture and paraspinal muscle characteristics for prediction of first vertebral fracture: A quantitative-CT analysis from the AGES cohort. Front. Endocrinol., 16.","DOI":"10.3389\/fendo.2025.1566424"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"e283","DOI":"10.1210\/clinem\/dgac722","article-title":"A CT-based radiomics nomogram for predicting osteoporotic vertebral fractures: A longitudinal study","volume":"108","author":"Wang","year":"2023","journal-title":"J. Clin. Endocrinol. Metab."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Zhang, D., Wang, J., Zhang, Y., Wei, Z., Wang, Y., Tang, H., Shen, J., Chen, X., and Xie, C. (2026). CT-based bone habitat radiomics for predicting risk of vertebral fracture in older adults: A longitudinal study. Glob. Spine J., 21925682261423096.","DOI":"10.1177\/21925682261423096"},{"key":"ref_24","first-page":"161","article-title":"The association between computed-tomography-based osteosarcopenia and osteoporotic vertebral fractures: A longitudinal study","volume":"48","author":"Tang","year":"2025","journal-title":"J. Endocrinol. Investig."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"3242","DOI":"10.1007\/s00586-024-08235-4","article-title":"Development and validation of a predictive model for vertebral fracture risk in osteoporosis patients","volume":"33","author":"Zhang","year":"2024","journal-title":"Eur. Spine J."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"24281","DOI":"10.1038\/s41598-024-75628-2","article-title":"Artificial intelligence in risk prediction and diagnosis of vertebral fractures","volume":"14","author":"Namireddy","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"111714","DOI":"10.1016\/j.ejrad.2024.111714","article-title":"Machine-learning value in the diagnosis of vertebral fractures: A systematic review and meta-analysis","volume":"180","author":"Li","year":"2024","journal-title":"Eur. J. Radiol."},{"key":"ref_28","first-page":"1841","article-title":"Predicting secondary vertebral compression fracture after vertebral augmentation via CT-based machine-learning radiomics-clinical model","volume":"32","author":"Wang","year":"2025","journal-title":"Acad. Radiol."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1521","DOI":"10.1007\/s00586-024-08579-x","article-title":"CT-based radiomics predicts adjacent vertebral fracture after percutaneous vertebral augmentation","volume":"34","author":"Yang","year":"2025","journal-title":"Eur. Spine J."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"e230614","DOI":"10.1148\/radiol.230614","article-title":"A CT-based deep-learning model for predicting subsequent fracture risk in hip-fracture patients","volume":"310","author":"Kim","year":"2024","journal-title":"Radiology"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"2207","DOI":"10.1007\/s00330-018-5846-8","article-title":"Vertebral body insufficiency fractures: Detection of vertebrae at risk on standard CT images using texture analysis and machine learning","volume":"29","author":"Muehlematter","year":"2019","journal-title":"Eur. Radiol."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Vabalas, A., Gowen, E., Poliakoff, E., and Casson, A.J. (2019). Machine-learning algorithm validation with a limited sample size. PLoS ONE, 14.","DOI":"10.1371\/journal.pone.0224365"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1137","DOI":"10.1002\/jbmr.5650080915","article-title":"Vertebral fracture assessment using a semiquantitative technique","volume":"8","author":"Genant","year":"1993","journal-title":"J. Bone Miner. Res."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1148\/radiol.2020191145","article-title":"The Image Biomarker Standardization Initiative: Standardized quantitative radiomics for high-throughput image-based phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1016\/j.bone.2008.10.048","article-title":"Vertebral fractures usually affect the cranial endplate because it is thinner and supported by less-dense trabecular bone","volume":"44","author":"Zhao","year":"2009","journal-title":"Bone"},{"key":"ref_36","first-page":"89","article-title":"Multicriteria conditional optimization based on genetic algorithms","volume":"2020","author":"Sineglazov","year":"2020","journal-title":"Syst. Res. Inf. Technol."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"e104","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational radiomics system to decode the radiographic phenotype","volume":"77","author":"Fedorov","year":"2017","journal-title":"Cancer Res."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"105800","DOI":"10.1016\/j.compbiomed.2022.105800","article-title":"Intelligent tuberculosis activity assessment system based on an ensemble of neural networks","volume":"147","author":"Sineglazov","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"ref_39","first-page":"1","article-title":"Classification and Analysis Topologies of Known Artificial Neurons and Neural Networks","volume":"904","author":"Zgurovsky","year":"2021","journal-title":"Stud. Comput. Intell."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"2887","DOI":"10.1016\/j.acra.2025.06.023","article-title":"Predicting thoracolumbar vertebral osteoporotic fractures: Value assessment of chest CT-based machine learning","volume":"32","author":"Chen","year":"2025","journal-title":"Acad. Radiol."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"33","DOI":"10.4103\/jcvjs.JCVJS_188_20","article-title":"Effectiveness and reliability of cannulated fenestrated screws augmented with polymethylmethacrylate cement in the surgical treatment of osteoporotic vertebral fractures","volume":"12","author":"Coniglio","year":"2021","journal-title":"J. Craniovertebr. Junction Spine"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Zuluaga-Garcia, J.P., Sierra, M.A., Call-Orellana, F.A., Herrera, D., Andrade-Almeida, R.A., Ravindran, P.K., and Ramirez-Ferrer, E. (2025). Complications of vertebroplasty in adults: Incidence, etiology, and therapeutic strategies\u2014A comprehensive, systematic literature review. Complications, 2.","DOI":"10.3390\/complications2030022"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"1484","DOI":"10.1007\/s00586-023-08026-3","article-title":"Clinical significance of cement leakage in kyphoplasty and vertebroplasty: A systematic review","volume":"33","author":"Rose","year":"2024","journal-title":"Eur. Spine J."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"264","DOI":"10.1186\/s13018-018-0952-5","article-title":"Balloon kyphoplasty versus percutaneous vertebroplasty for osteoporotic vertebral compression fracture: A meta-analysis and systematic review","volume":"13","author":"Wang","year":"2018","journal-title":"J. Orthop. Surg. Res."}],"container-title":["Medicina"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1648-9144\/62\/8\/1518\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,8]],"date-time":"2026-08-08T04:13:36Z","timestamp":1786162416000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1648-9144\/62\/8\/1518"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,6]]},"references-count":44,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["medicina62081518"],"URL":"https:\/\/doi.org\/10.3390\/medicina62081518","relation":{},"ISSN":["1648-9144"],"issn-type":[{"value":"1648-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,6]]}}}