{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T16:57:31Z","timestamp":1785430651954,"version":"3.56.0"},"reference-count":34,"publisher":"MDPI AG","issue":"13","license":[{"start":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T00:00:00Z","timestamp":1783036800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100011093","name":"University of Phayao","doi-asserted-by":"publisher","award":["2253\/2568"],"award-info":[{"award-number":["2253\/2568"]}],"id":[{"id":"10.13039\/501100011093","id-type":"DOI","asserted-by":"publisher"}]},{"award":["2253\/2568"],"award-info":[{"award-number":["2253\/2568"]}],"id":[{"id":"https:\/\/ror.org\/00a5mh069","id-type":"ROR","asserted-by":"publisher"}]},{"name":"School of Medicine, University of Phayao"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JCM"],"abstract":"<jats:p>Background\/Objectives: Osteoporosis is a major public health concern associated with increased fracture risk and reduced quality of life if not detected at an early stage. Automated analysis of knee X-ray images using artificial intelligence has shown promising potential for opportunistic osteoporosis screening. This study aims to develop and evaluate a multi-stage deep learning and machine learning framework for osteoporosis classification, with particular emphasis on external validation, calibration drift, and cross-domain generalization performance. Methods: Knee X-ray images were categorized into three classes: Normal, Osteopenia, and Osteoporosis. Deep features were extracted using pretrained convolutional neural networks, including ResNet18, EfficientNetB0, and DenseNet121. The extracted features were subsequently classified using multiple machine learning models, including Neural Network, Efficient Linear, Support Vector Machine, and Naive Bayes classifiers. Two data augmentation strategies were investigated: targeted minority-class augmentation and full 3\u00d7 dataset expansion with class balancing. Model performance was evaluated using accuracy, precision, recall, F1-score, and AUC on internal validation, independent test sets, and external validation datasets. Additional analyses included reliability calibration assessment, isotonic recalibration, and class-prior boosting with cross-validated threshold optimization to address external domain shift. Results: EfficientNetB0 and DenseNet121 consistently outperformed ResNet18 across most evaluation metrics. Under the balanced augmentation strategy, EfficientNetB0 combined with Efficient Linear demonstrated strong and stable performance, while DenseNet121 paired with a Neural Network achieved the highest overall classification performance. External validation revealed a substantial discrepancy between AUC and threshold-based metrics, indicating the presence of calibration drift and class-prior mismatch across imaging domains. Reliability analysis showed severe probability collapse in the Osteopenia class during external testing. Post-hoc recalibration improved probability reliability, while class-prior boosting substantially increased Osteopenia sensitivity and improved balanced accuracy and macro F1-score under external validation conditions. Conclusions: The proposed framework demonstrates the feasibility of combining pretrained CNN-based deep feature extraction with machine learning classifiers for osteoporosis classification from knee X-ray images. The findings further highlight that maintaining model performance under external testing conditions may require not only strong feature extraction capability but also adaptive recalibration and deployment-aware threshold optimization to address calibration drift and cross-domain variability. While the results are encouraging, the present study should be considered a proof-of-concept investigation. Although the framework was evaluated using an independent public external dataset, further validation using larger and more diverse multi-center clinical cohorts is necessary to establish generalizability and clinical utility before routine clinical implementation can be considered.<\/jats:p>","DOI":"10.3390\/jcm15135222","type":"journal-article","created":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T14:33:06Z","timestamp":1783089186000},"page":"5222","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Opportunistic Osteoporosis Screening from Routine Knee Radiographs Using a Multi-Stage CNN Framework with External Validation"],"prefix":"10.3390","volume":"15","author":[{"given":"Nitiphoom","family":"Sinnathakorn","sequence":"first","affiliation":[{"name":"School of Medicine, University of Phayao, Phayao 56000, Thailand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chanon","family":"Fahpinyo","sequence":"additional","affiliation":[{"name":"School of Medicine, University of Phayao, Phayao 56000, Thailand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8563-017X","authenticated-orcid":false,"given":"Watcharaporn","family":"Cholamjiak","sequence":"additional","affiliation":[{"name":"School of Science, University of Phayao, Phayao 56000, Thailand"},{"name":"MedUP Intelligence Healthcare and Data Innovation Center, School of Medicine, University of Phayao, Phayao 56000, Thailand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Suthep","family":"Suantai","sequence":"additional","affiliation":[{"name":"Research Center in Optimization and Computational Intelligence for Big Data Prediction, Department of Mathematics, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,3]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s00198-018-4704-5","article-title":"European guidance for the diagnosis and management of osteoporosis in postmenopausal women","volume":"30","author":"Kanis","year":"2019","journal-title":"Osteopor. Int."},{"key":"ref_2","first-page":"46","article-title":"An overview and management of osteoporosis","volume":"4","year":"2016","journal-title":"Eur. J. Rheumatol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1007\/s11657-014-0182-3","article-title":"Estimating prevalence of osteoporosis: Examples from industrialized countries","volume":"9","author":"Wade","year":"2014","journal-title":"Arch. Osteoporos."},{"key":"ref_4","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":"Osteopor. Int."},{"key":"ref_5","unstructured":"World Health Organization (2004). WHO Scientific Group on the Assessment of Osteoporosis at Primary Health Care Level, WHO Press."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1136\/pgmj.2007.057505","article-title":"The role of DXA bone density scans in the diagnosis and treatment of osteoporosis","volume":"83","author":"Blake","year":"2007","journal-title":"Postgrad. Med. J."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Valero-Mart\u00ednez, C., Castillo-Morales, V., G\u00f3mez-Le\u00f3n, N., Hern\u00e1ndez-P\u00e9rez, I., Vicente-Rabaneda, E.F., Uriarte, M., and Casta\u00f1eda, S. (2024). Application of nuclear medicine techniques in musculoskeletal infection: Current trends and future prospects. J. Clin. Med., 13.","DOI":"10.3390\/jcm13041058"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"676","DOI":"10.1177\/08465371231164743","article-title":"Machine learning for opportunistic screening for osteoporosis and osteopenia using knee CT scans","volume":"74","author":"Sebro","year":"2023","journal-title":"Can. Assoc. Radiol. J."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e75965","DOI":"10.2196\/75965","article-title":"The diagnostic value of image-based machine learning for osteoporosis: Systematic review and meta-analysis","volume":"28","author":"Zhao","year":"2026","journal-title":"J. Med. Internet Res."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"196","DOI":"10.5114\/reum\/190262","article-title":"Magnetic resonance imaging of the musculoskeletal system in the diagnosis of rheumatic diseases in the pediatric population","volume":"62","author":"Kmiecik","year":"2024","journal-title":"Reumatologia"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med. Image Anal."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"25014","DOI":"10.1038\/s41598-024-73709-w","article-title":"Automated deep learning-based bone mineral density assessment for opportunistic osteoporosis screening using various CT protocols with multi-vendor scanners","volume":"14","author":"Park","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"D\u2019Agostino, V., Sorriento, A., Cafarelli, A., Donati, D., Papalexis, N., Russo, A., Lisignoli, G., Ricotti, L., and Spinnato, P. (2024). Ultrasound imaging in knee osteoarthritis: Current role, recent advancements, and future perspectives. J. Clin. Med., 13.","DOI":"10.3390\/jcm13164930"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2530003","DOI":"10.1142\/S0219649225300037","article-title":"A systematic review and meta-analysis on diagnosis of osteoporosis risk in real time using machine learning and deep learning techniques","volume":"25","author":"Gomathi","year":"2026","journal-title":"J. Inf. Knowl. Manag."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1093\/dmfr\/twaf052","article-title":"Deep learning for osteoporosis screening in dental practice: A systematic review","volume":"55","author":"Querrer","year":"2026","journal-title":"Dentomaxillofac. Radiol."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 27\u201330 June 2016, IEEE.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_17","unstructured":"Tan, M., and Le, Q. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the International Conference on Machine Learning (ICML), Long Beach, CA, USA, 9\u201315 June 2019, PMLR."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 22\u201329 October 2017, IEEE.","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"455","DOI":"10.1016\/j.arthro.2024.01.043","article-title":"Medical imaging applications developed using artificial intelligence demonstrate high internal validity yet are limited in scope and lack external validation","volume":"41","author":"Oeding","year":"2025","journal-title":"Arthrosc. J. Arthrosc. Relat. Surg."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Kelly, C.J., Karthikesalingam, A., Suleyman, M., Corrado, G., and King, D. (2019). Key challenges for delivering clinical impact with artificial intelligence. BMC Med., 17.","DOI":"10.1186\/s12916-019-1426-2"},{"key":"ref_21","first-page":"210","article-title":"Machine Learning-Based Prediction and Risk Factor Analysis of Osteoporosis","volume":"2","author":"Waratamrongpatai","year":"2025","journal-title":"J. Decis. Mak. Healthc."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"117176","DOI":"10.1016\/j.bone.2024.117176","article-title":"AI-based opportunistic quantitative image analysis of lung cancer screening CTs to reduce disparities in osteoporosis screening","volume":"186","author":"Huber","year":"2024","journal-title":"Bone"},{"key":"ref_23","unstructured":"Gobara, M. (2025, November 01). Multi-Class Knee Osteoporosis X-Ray Dataset. Available online: https:\/\/www.kaggle.com\/datasets\/mohamedgobara\/multi-class-knee-osteoporosis-x-ray-dataset."},{"key":"ref_24","first-page":"521","article-title":"Multi-model attention-enhanced CNN ensemble for automated osteoporosis detection in radiographic knee images","volume":"18","author":"Ichahane","year":"2025","journal-title":"Int. J. Intell. Eng. Syst."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"4444","DOI":"10.1038\/s41467-025-59433-7","article-title":"A clinically applicable AI system for detection and diagnosis of bone metastases using CT scans","volume":"16","author":"Zhang","year":"2025","journal-title":"Nat. Commun."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e202500070","DOI":"10.1002\/aidi.202500070","article-title":"Artificial intelligence for bone: Theory, methods, and applications","volume":"2","author":"Yuan","year":"2025","journal-title":"Adv. Intell. Discov."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Huang, G., Liu, Z., Van Der Maaten, L., and Weinberger, K.Q. (2017). Densely connected convolutional networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 21\u201326 July 2017, IEEE.","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref_28","unstructured":"MathWorks (2025). Classification Learner App; MATLAB R2025a Documentation, The MathWorks, Inc.. Available online: https:\/\/www.mathworks.com\/help\/stats\/classificationlearner-app.html."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Breiman, L., Friedman, J., Olshen, R.A., and Stone, C.J. (2017). Classification and Regression Trees, Chapman and Hall\/CRC.","DOI":"10.1201\/9781315139470"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_31","unstructured":"Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, MIT Press."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern classification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_33","unstructured":"Haykin, S. (2009). Neural Networks and Learning Machines, Pearson. [3rd ed.]."},{"key":"ref_34","first-page":"1871","article-title":"LIBLINEAR: A library for large linear classification","volume":"9","author":"Fan","year":"2008","journal-title":"J. Mach. Learn. Res."}],"container-title":["Journal of Clinical Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/13\/5222\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T12:06:22Z","timestamp":1783685182000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/13\/5222"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,3]]},"references-count":34,"journal-issue":{"issue":"13","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["jcm15135222"],"URL":"https:\/\/doi.org\/10.3390\/jcm15135222","relation":{},"ISSN":["2077-0383"],"issn-type":[{"value":"2077-0383","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,3]]}}}