{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T05:55:33Z","timestamp":1790920533006,"version":"4.1.0"},"reference-count":58,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T00:00:00Z","timestamp":1790899200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Osteoporosis is a major global health burden, with considerable morbidity, mortality, and healthcare costs. Artificial intelligence (AI) has opened new avenues for osteoporosis screening, bone mineral density quantification, fracture-risk prediction, and clinical decision support, but progress remains fragmented across machine learning, deep learning, explainable AI (XAI), generative models, and multimodal learning. This systematic review brings these strands together and identifies priority research needs for AI-based osteoporosis prediction. Following the PRISMA 2020 statement, studies were identified from Scopus and Google Scholar, published between January 2020 and the final search date of 30 June 2026, with two reviewers screening independently and resolving disagreements by consensus. Design-related risk-of-bias indicators, such as retrospective vs. prospective design, validation strategy, and reference standard, were recorded for every study; a full PROBAST, QUADAS-2, and Newcastle-Ottawa Scale appraisal is planned but not yet complete. Fifty-eight studies met the inclusion criteria and were synthesized narratively and descriptively, without formal meta-analysis, given substantial heterogeneity. Findings spanned five themes: classical machine learning, deep learning for imaging, explainable AI, generative models, and multimodal learning. Deep learning achieved strong predictive accuracy (AUC 0.82\u20130.96), XAI improved clinical interpretability, generative models helped offset data scarcity, and multimodal approaches boosted performance by combining imaging and clinical data. Key gaps remain, including limited external validation, heavy reliance on retrospective data, poor generalisability, and limited regulatory readiness. Building on these findings, this review proposes MXGNet, a multimodal, explainable, generative framework incorporating privacy-preserving federated learning, intended to guide development of robust, interpretable, clinically deployable AI tools for osteoporosis prediction.<\/jats:p>","DOI":"10.3389\/frai.2026.1930437","type":"journal-article","created":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T05:35:34Z","timestamp":1790919334000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Artificial intelligence, explainable deep learning, and generative models for osteoporosis and bone disease prediction: a systematic review and future research roadmap"],"prefix":"10.3389","volume":"9","author":[{"given":"M.","family":"Raja","sequence":"first","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore","place":["Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Avulapalli","family":"Jayaram Reddy","sequence":"additional","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore","place":["Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,2]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"51","DOI":"10.1007\/s10916-026-02373-6","article-title":"SHIELD: an AI framework for skeletal health intelligence and early lesion detection to improve orthopedic referrals","volume":"50","author":"Alili","year":"2026","journal-title":"J. Med. Syst"},{"key":"B2","doi-asserted-by":"publisher","first-page":"e22550","DOI":"10.2196\/22550","article-title":"Deep learning with electronic health records for short-term fracture risk identification: crystal bone algorithm development and validation","volume":"22","author":"Almog","year":"2020","journal-title":"J. Med. Internet Res"},{"key":"B3","doi-asserted-by":"publisher","first-page":"215","DOI":"10.14445\/23488549\/IJECE-V12I6P117","article-title":"A novel deep learning approach with attention mechanism for early osteoporosis detection from knee X-ray imaging","volume":"12","author":"Athira","year":"2025","journal-title":"SSRG Int. J. Electron. Commun. Eng"},{"key":"B4","doi-asserted-by":"publisher","first-page":"199","DOI":"10.22266\/ijies2025.0731.13","article-title":"Advanced osteoporosis prediction from knee X-rays via residual convolution and recurrent networks","volume":"18","author":"Athira","year":"2025","journal-title":"Int. J. Intell. Eng. Syst"},{"key":"B5","doi-asserted-by":"publisher","first-page":"3381","DOI":"10.1007\/s10489-022-03682-2","article-title":"An innovative medical image synthesis based on dual gan deep neural networks for improved segmentation quality: an innovative medical image synthesis based on dual gan deep neural networks for improved segmentation quality","volume":"53","author":"Beji","year":"2023","journal-title":"Appl. Intell"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1318","DOI":"10.1038\/s41586-026-10181-8","article-title":"Merlin: a computed tomography vision-language foundation model and dataset","volume":"652","author":"Blankemeier","year":"2026","journal-title":"Nature"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1995","DOI":"10.1007\/s11263-023-01791-0","article-title":"Autoencoder-driven multimodal collaborative learning for medical image synthesis","volume":"131","author":"Cao","year":"2023","journal-title":"Int. J. Comput. Vis"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1538477","DOI":"10.3389\/fdgth.2025.1538477","article-title":"A labeled dataset for osteoporosis screening based on electromagnetic attenuation","volume":"7","author":"Carvalho","year":"2025","journal-title":"Front. Digit. Health"},{"key":"B9","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1038\/s41746-026-02484-x","article-title":"Advancing diagnostic equity through artificial intelligence chest radiograph screening for osteoporosis in Asian populations","volume":"9","author":"Chen","year":"2026","journal-title":"NPJ Digit. Med"},{"key":"B10","doi-asserted-by":"publisher","first-page":"3295","DOI":"10.1007\/s10489-023-05017-1","article-title":"A general variation-driven network for medical image synthesis","volume":"54","author":"Chen","year":"2024","journal-title":"Appl. Intell"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1554","DOI":"10.14569\/IJACSA.2024.01506154","article-title":"Bone quality classification of dual energy X-ray absorptiometry images using convolutional neural network models","volume":"15","author":"Gonzalez","year":"2024","journal-title":"Int. J. Adv. Comput. Sci. Appl"},{"key":"B12","doi-asserted-by":"publisher","first-page":"102970","DOI":"10.1016\/j.media.2023.102970","article-title":"Bone mineral density estimation from a plain X-ray image by learning decomposition into projections of bone-segmented computed tomography","volume":"90","author":"Gu","year":"2023","journal-title":"Med. Image Anal"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1582960","DOI":"10.3389\/frai.2025.1582960","article-title":"A comparative study of bone density in elderly people measured with ai and qct","volume":"8","author":"Guo","year":"2025","journal-title":"Front. Artif. Intell"},{"key":"B14","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1038\/s41746-021-00519-z","article-title":"Clinical knowledge extraction via sparse embedding regression (KESER) with multi-center large scale electronic health record data","volume":"4","author":"Hong","year":"2021","journal-title":"NPJ Digit. Med"},{"key":"B15","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1007\/s10916-025-02333-6","article-title":"Artificial intelligence-enabled electrocardiography identifies osteoporosis and has prognostic value","volume":"49","author":"Hsing","year":"2025","journal-title":"J. Med. Syst"},{"key":"B16","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1186\/s12891-025-09389-y","article-title":"Performance of clinical prediction models for identifying postmenopausal osteoporosis: a systematic review and meta-analysis","volume":"27","author":"Huang","year":"2026","journal-title":"BMC Musculoskelet. Disord"},{"key":"B17","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1177\/08465371251380240","article-title":"AI screening tool based on X-rays improves early detection of decreased bone density in a clinical setting","volume":"77","author":"Jayarajah","year":"2026","journal-title":"Can. Assoc. Radiol. J"},{"key":"B18","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1117\/12.2612893","article-title":"\u201cEvaluating procedures for establishing generative adversarial network-based stochastic image models in medical imaging,\u201d","author":"Kelkar","year":"2022","journal-title":"Medical imaging 2022: image perception, observer performance, and technology assessment"},{"key":"B19","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1007\/s11914-026-00961-6","article-title":"Advances in imaging-based fracture risk assessment for unlocking latent skeletal fragility","volume":"24","author":"Kim","year":"2026","journal-title":"Curr. Osteoporos. Rep"},{"key":"B20","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1038\/s41746-026-02363-5","article-title":"Large language models improve transferability of electronic health record-based predictions across countries and coding systems","volume":"9","author":"Kirchler","year":"2026","journal-title":"NPJ Digit. Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1038\/s41746-020-0301-z","article-title":"Deep representation learning of electronic health records to unlock patient stratification at scale","volume":"3","author":"Landi","year":"2020","journal-title":"NPJ Digit. Med"},{"key":"B22","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1186\/s12891-025-09385-2","article-title":"Risk prediction models for postmenopausal osteoporosis: a systematic review and meta-analysis study","volume":"27","author":"Li","year":"2026","journal-title":"BMC Musculoskelet. Disord"},{"key":"B23","doi-asserted-by":"publisher","first-page":"2616","DOI":"10.1109\/TMI.2023.3261822","article-title":"Low-dose CT image synthesis for domain adaptation imaging using a generative adversarial network with noise encoding transfer learning","volume":"42","author":"Li","year":"2023","journal-title":"IEEE Trans. Med. Imaging"},{"key":"B24","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1186\/s12891-026-09526-1","article-title":"Elucidating osteoporosis response signatures in rheumatoid arthritis using explainable machine learning ensembles","volume":"27","author":"Lin","year":"2026","journal-title":"BMC Musculoskelet. Disord"},{"key":"B25","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1186\/s12859-022-04596-z","article-title":"A hierarchical opportunistic screening model for osteoporosis using machine learning applied to clinical data and CT images","volume":"23","author":"Liu","year":"2022","journal-title":"BMC Bioinform"},{"key":"B26","doi-asserted-by":"publisher","first-page":"20552076241257456","DOI":"10.1177\/20552076241257456","article-title":"Predicting fracture risk for elderly osteoporosis patients by hybrid machine learning model","volume":"10","author":"Liu","year":"2024","journal-title":"Digit. Health"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1394","DOI":"10.1007\/s10278-025-01547-x","article-title":"Deep learning for osteoporosis diagnosis using magnetic resonance images of lumbar vertebrae","volume":"39","author":"Mousavinasab","year":"2025","journal-title":"J. Imaging Inform. Med"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1007\/s11831-026-10629-0","article-title":"Federated learning in multimodal healthcare diagnostics: privacy-preserving AI for biomedical imaging, electronic health records, wearables, and clinical decision support","author":"Nemane","year":"2026","journal-title":"Arch. Comput. Methods Eng"},{"key":"B29","doi-asserted-by":"publisher","first-page":"4255","DOI":"10.1007\/s10278-025-01469-8","article-title":"SADiff: a sinogram-aware diffusion model for low-dose CT image denoising","volume":"38","author":"Niknejad Mazandarani","year":"2025","journal-title":"J. Imaging Inform. Med"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.3823","article-title":"A stable diffusion-based framework for multi-class stroke lesion segmentation in non-contrast brain ct","author":"'\u0301Ozcan","year":"2026","journal-title":"PeerJ Computer Science"},{"key":"B31","doi-asserted-by":"publisher","first-page":"e3577","DOI":"10.7717\/peerj-cs.3577","article-title":"Prompt-guided synthetic data generation to mitigate domain shift in periapical lesion detection","volume":"12","author":"'\u0301Oz'\u0301ud'\u0301o\u011fru","year":"2026","journal-title":"PeerJ Comput. Sci."},{"key":"B32","doi-asserted-by":"crossref","DOI":"10.1109\/APCIT65661.2025.11410996","article-title":"\u201cExplainable AI for osteoporosis risk modeling: random forest and SHAP interpretation on CTGAN-enhanced data,\u201d","volume-title":"2025 2nd Asia Pacific conference on innovation in technology (APCIT)","author":"Padmanabha","year":"2025"},{"key":"B33","doi-asserted-by":"publisher","first-page":"399","DOI":"10.22266\/ijies2025.1130.26","article-title":"Osteonet: a hybrid deep learning model with DenseNet 121-CBAM for automated osteoporosis detection in knee X-rays","volume":"18","author":"Ramachandran","year":"2025","journal-title":"Int. J. Intell. Eng. Syst"},{"key":"B34","doi-asserted-by":"publisher","first-page":"492","DOI":"10.2991\/978-94-6239-654-8_40","article-title":"\u201cA multi-modal approach for automated osteoporosis diagnosis using knee X-ray imaging data,\u201d","author":"Ramamoorthy","year":"2026","journal-title":"Global conference on sustainable energy systems, smart electronics and intelligent computing (GCSESEIC 2025)"},{"key":"B35","doi-asserted-by":"publisher","first-page":"395","DOI":"10.3233\/IDT-240227","article-title":"Osteoporosis classification using VGG-16 and logistic regression: a radiograph and bmd data approach","volume":"19","author":"Salunke","year":"2025","journal-title":"Intell. Decis. Technol"},{"key":"B36","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1007\/s44196-024-00615-4","article-title":"Knee osteoporosis diagnosis based on deep learning","volume":"17","author":"Sarhan","year":"2024","journal-title":"Int. J. Comput. Intell. Syst"},{"key":"B37","doi-asserted-by":"publisher","first-page":"2261","DOI":"10.1007\/s11548-023-02910-5","article-title":"Detecting whether l1 or other lumbar levels would be excluded from DXA bone mineral density analysis during opportunistic CT screening for osteoporosis using machine learning","volume":"18","author":"Sebro","year":"2023","journal-title":"Int. J. Comput. Assist. Radiol. Surg"},{"key":"B38","doi-asserted-by":"publisher","first-page":"2631","DOI":"10.1007\/s00261-025-05213-2","article-title":"Advancing osteoporosis opportunistic screening: multicenter validation of a deep learning algorithm using abdominal CT scans","volume":"51","author":"Serpa","year":"2026","journal-title":"Abdom. Radiol"},{"key":"B39","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s12178-025-09997-0","article-title":"Artificial intelligence applications in musculoskeletal imaging","volume":"19","author":"Shariatnia","year":"2026","journal-title":"Curr. Rev. Musculoskelet. Med"},{"key":"B40","doi-asserted-by":"publisher","first-page":"394","DOI":"10.1186\/s12891-026-09759-0","article-title":"Machine learning-based risk predictive model for postoperative refractures in patients with osteoporotic vertebral compression fractures: a systematic review and critical appraisal","volume":"28","author":"Shi","year":"2026","journal-title":"BMC Musculoskelet. Disord"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1351","DOI":"10.1007\/s00256-026-05147-w","article-title":"Vertebral-level discrimination of incidental vertebral fractures using volumetric bmd, texture features, and finite element-derived fracture load: an exploratory study","volume":"55","author":"Strack","year":"2026","journal-title":"Skeletal Radiol"},{"key":"B42","doi-asserted-by":"publisher","first-page":"302","DOI":"10.1038\/s41746-025-01712-0","article-title":"Multimodal fusion model for prognostic prediction and radiotherapy response assessment in head and neck squamous cell carcinoma","volume":"8","author":"Tian","year":"2025","journal-title":"NPJ Digit. Med"},{"key":"B43","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1007\/s10916-023-02030-2","article-title":"Artificial intelligence-enabled chest X-ray classifies osteoporosis and identifies mortality risk","volume":"48","author":"Tsai","year":"2024","journal-title":"J. Med. Syst"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.1186\/s12866-026-04718-0","article-title":"Machine learning model integrating oral microbiota and clinical features for predicting osteoporosis and bone loss in high-altitude populations","author":"Wang","year":"2026","journal-title":"BMC Microbiol"},{"key":"B45","doi-asserted-by":"publisher","first-page":"160","DOI":"10.1186\/s12911-026-03350-2","article-title":"An intelligent auxiliary diagnostic system for early osteoporosis screening using stacking ensemble learning","volume":"26","author":"Wang","year":"2026","journal-title":"BMC Med. Inform. Decis. Mak"},{"key":"B46","doi-asserted-by":"publisher","first-page":"981","DOI":"10.1007\/s00256-026-05185-4","article-title":"Artificial intelligence enabled prediction of hip fragility fracture risk using hip radiograph: an attractive approach yet more work needs to be done","volume":"55","author":"W\u00e1ng","year":"2026","journal-title":"Skeletal Radiol"},{"key":"B47","doi-asserted-by":"publisher","first-page":"14193","DOI":"10.1007\/s11042-022-13911-y","article-title":"Osteoporosis diagnosis in knee X-rays by transfer learning based on convolution neural network","volume":"82","author":"Wani","year":"2023","journal-title":"Multimed. Tools Appl"},{"key":"B48","doi-asserted-by":"publisher","first-page":"21041","DOI":"10.1007\/s00521-024-10158-6","article-title":"Deep ensemble learning for osteoporosis diagnosis from knee X-rays: a preliminary cohort study in Kashmir Valley","volume":"36","author":"Wani","year":"2024","journal-title":"Neural Comput. Appl"},{"key":"B49","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1007\/s10462-025-11440-2","article-title":"Review of generative AI for synthetic data generation: a healthcare perspective","volume":"59","author":"Waseem","year":"2026","journal-title":"Artif. Intell. Rev"},{"key":"B50","doi-asserted-by":"publisher","first-page":"248","DOI":"10.1007\/s00371-026-04438-6","article-title":"CIPHER: cross-modal integrated and pathology-aware embedded representation diffusion model with temporal consistency for brain MRI-to-PET translation","volume":"42","author":"Wu","year":"2026","journal-title":"Vis. Comput"},{"key":"B51","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1007\/s11914-026-00953-6","article-title":"Deep learning for analysis of bone marrow adiposity: breakthroughs from recent large-scale analyses in the UK Biobank","volume":"24","author":"Xu","year":"2026","journal-title":"Curr. Osteoporos. Rep"},{"key":"B52","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1007\/s00011-026-02259-w","article-title":"ABCA8 drives bone loss via disrupting Th17\/Treg imbalance: a novel immunometabolic target for osteoporosis","volume":"75","author":"Yi","year":"2026","journal-title":"Inflamm. Res"},{"key":"B53","doi-asserted-by":"publisher","first-page":"2428","DOI":"10.1007\/s10278-024-01135-5","article-title":"An automated vertebrae localization, segmentation, and osteoporotic compression fracture detection pipeline for computed tomographic imaging","volume":"37","author":"Y\u0131ld\u0131z Potter","year":"2024","journal-title":"J. Imaging Inform. Med"},{"key":"B54","doi-asserted-by":"publisher","first-page":"7299","DOI":"10.1109\/JBHI.2025.3572458","article-title":"HealthiVert-GAN: a novel framework of pseudo-healthy vertebral image synthesis for interpretable compression fracture grading","volume":"29","author":"Zhang","year":"2025","journal-title":"IEEE J. Biomed. Health Inform"},{"key":"B55","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s43926-025-00233-2","article-title":"StyleGAN based path analysis from image to video generation in multimedia generation","volume":"5","author":"Zhang","year":"2025","journal-title":"Discov Internet Things"},{"key":"B56","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1186\/s13040-026-00520-w","article-title":"An online non-radiographic osteoporosis prediction calculator constructed using interpretable machine learning","volume":"19","author":"Zhang","year":"2026","journal-title":"BioData Min"},{"key":"B57","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1186\/s12891-024-07309-0","article-title":"Comparative evaluation of multiparametric lumbar MRI radiomic models for detecting osteoporosis","volume":"25","author":"Zhen","year":"2024","journal-title":"BMC Musculoskelet. Disord"},{"key":"B58","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1007\/s11914-026-00950-9","article-title":"Current status of AI-assisted screening for opportunistic osteoporosis","volume":"24","author":"Zheng","year":"2026","journal-title":"Curr. Osteoporos. Rep"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1930437\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T05:35:35Z","timestamp":1790919335000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1930437\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,2]]},"references-count":58,"alternative-id":["10.3389\/frai.2026.1930437"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1930437","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,2]]},"article-number":"1930437"}}