{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,12]],"date-time":"2026-04-12T16:57:32Z","timestamp":1776013052010,"version":"3.50.1"},"reference-count":40,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T00:00:00Z","timestamp":1753228800000},"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:sec><jats:title>Introduction<\/jats:title><jats:p>Knee injuries, especially Anterior Cruciate Ligament (ACL) tears and meniscus tears, are becoming increasingly common and can severely restrict mobility and quality of life. Early diagnosis is essential for effective treatment and for preventing long-term complications such as knee osteoarthritis. While deep learning approaches have shown promise in identifying knee injuries from MRI scans, they often require large amounts of labeled data, which can be both scarce and privacy-sensitive.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>This paper analyses a hybrid methodology that integrates few-shot learning with federated learning for the diagnosis of knee injuries using MRI scans. The proposed model used a 3DResNet50 architecture as the backbone to enhance both feature extraction and embedding representation. A combined Centralized and Federated Few-Shot Learning Framework is analysed to leverage episodic-intermittent training strategy based on Prototypical Networks. The model is trained incorporating Stochastic Gradient Descent (SGD), Cross-Entropy Loss, and a MultiStep Learning Rate scheduler to enhance few-shot classification. This model also addressed the challenge of limited annotated data ensuring patient data privacy through distributed learning across multiple regions.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The models performance was evaluated on the MRNet dataset for multi-label classification. In the centralized setting, the model achieved accuracies of 85.3% on axial views, 82.1% on sagittal views, and 71% on coronal views. The propose work attained accuracies as 83% (axial), 83.9% (sagittal), and 65% (coronal), demonstrating the framework\u2019s effectiveness across different learning configurations.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>The proposed method outperforms in diagnostic accuracy, generalization across MRI planes, and patient privacy via federated learning. However, it faces limitations, including lower coronal view performance and high computational demands due to its complex architecture.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1589358","type":"journal-article","created":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T14:04:26Z","timestamp":1753279466000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Federated knee injury diagnosis using few shot learning"],"prefix":"10.3389","volume":"8","author":[{"given":"Chirag","family":"Goel","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anita","family":"X","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jani Anbarasi","family":"L","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,7,23]]},"reference":[{"key":"ref1","first-page":"1","article-title":"Knee injury diagnosis with data and feature fusion-enhanced multi-label classification network","author":"Al Ghothani","year":"2023"},{"key":"ref2","first-page":"149","article-title":"A comparative study of existing and new deep learning methods for detecting knee injuries using the MRNet dataset","author":"Azcona","year":"2020"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"136","DOI":"10.1080\/03008207.2018.1470167","article-title":"Conditioned media from human osteoarthritic synovium induces inflammation in a synoviocyte cell line","volume":"60","author":"Belluzzi","year":"2019","journal-title":"Connect. Tissue Res."},{"key":"ref4","first-page":"71","author":"Belton","year":"2021"},{"key":"ref5","article-title":"Easyfsl","author":"Bennequin","year":"2023"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"e1002699","DOI":"10.1371\/journal.pmed.1002699","article-title":"Deep-learning-assisted diagnosis for knee magnetic resonance imaging: development and retrospective validation of MRNet","volume":"15","author":"Bien","year":"2018","journal-title":"PLoS Med."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"1149","DOI":"10.1111\/os.12577","article-title":"Challenging surgical treatment of displaced articular Tibial plateau fractures: do early knee radiographic features have a predictive value of the mid-term clinical functional outcomes?","volume":"11","author":"Biz","year":"2019","journal-title":"Orthop. Surg."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1110","DOI":"10.3390\/ijerph19031110","article-title":"Are patellofemoral ligaments and retinacula distinct structures of the knee joint? An anatomic, histological and magnetic resonance imaging study","volume":"19","author":"Biz","year":"2022","journal-title":"Int. J. Environ. Res. Public Health"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"980","DOI":"10.1007\/s10278-019-00193-4","article-title":"Deep learning for detection of complete anterior cruciate ligament tear","volume":"32","author":"Chang","year":"2019","journal-title":"J. Digit. Imaging"},{"key":"ref10","article-title":"Dfml: dynamic federated meta-learning for rare disease prediction","author":"Chen","year":"2023","journal-title":"IEEE\/ACM Trans. Comput. Biol. Bioinform."},{"key":"ref11","first-page":"706","article-title":"Metalr: Meta-tuning of learning rates for transfer learning in medical imaging","author":"Chen","year":"2023"},{"key":"ref12","first-page":"1","article-title":"Med3d: transfer learning for 3d medical image analysis","author":"Chen","year":"2019"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"1384","DOI":"10.3390\/diagnostics11081384","article-title":"Transmed: transformers advance multi-modal medical image classification","volume":"11","author":"Dai","year":"2021","journal-title":"Diagnostics"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"102142","DOI":"10.1016\/j.compmedimag.2022.102142","article-title":"Deep convolutional feature details for better knee disorder diagnoses in magnetic resonance images","volume":"102","author":"Dunnhofer","year":"2022","journal-title":"Comput- erized Med. Imaging Graphics"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"740","DOI":"10.1002\/jmri.28284","article-title":"Automatic detection of meniscus tears using backbone convolu- tional neural networks on knee MRI","volume":"57","author":"Hung","year":"2023","journal-title":"J. Magn. Reson. Imaging"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"3779","DOI":"10.1007\/s11063-022-11014-1","article-title":"Effectiveness of federated learning and CNN ensemble architectures for identifying brain tumors using MRI images","volume":"55","author":"Islam","year":"2023","journal-title":"Neural. Process. Lett."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"105","DOI":"10.3390\/diagnostics11010105","article-title":"Efficient detection of knee ante- rior cruciate ligament from magnetic resonance imaging using deep learning ap- proach","volume":"11","author":"Javed Awan","year":"2021","journal-title":"Diagnostics"},{"key":"ref18","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1109\/JBHI.2022.3215147","article-title":"Multi-learner based deep meta-learning for few-shot medical image classification","volume":"27","author":"Jiang","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"2314","DOI":"10.3390\/diagnostics12102314","article-title":"Anterior cruciate ligament tear detection based on deep convolutional neural network","volume":"12","author":"Joshi","year":"2022","journal-title":"Diagnostics"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1009","DOI":"10.3390\/make3040050","article-title":"Detection and classification of knee injuries from MR images using the MRNet dataset with progressively operating deep learn- ing methods","volume":"3","author":"Kara","year":"2021","journal-title":"Mach. Learn. Knowl. Extr."},{"key":"ref21","author":"Khandelwal","year":"2020"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"3651","DOI":"10.1109\/TMI.2023.3300725","article-title":"Federated domain adaptation via transformer for multi-site Alzheimer\u2019s disease diagnosis","volume":"42","author":"Lei","year":"2023","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"105144","DOI":"10.1016\/j.compbiomed.2021.105144","article-title":"Domain generalization on medical imaging classification using episodic training with task augmentation","volume":"141","author":"Li","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1016\/j.jot.2022.05.006","article-title":"Identification and diagnosis of meniscus tear by magnetic resonance imaging using a deep learning model","volume":"34","author":"Li","year":"2022","journal-title":"J. Orthop. Transl."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"1268543","DOI":"10.3389\/fbioe.2023.1268543","article-title":"Automated diagnosis of anterior cruciate ligament via a weighted multi-view network","volume":"11","author":"Li","year":"2023","journal-title":"Front. Bioeng. Biotechnol."},{"key":"ref26","first-page":"100409","article-title":"Trends in Australian knee injury rates: an epidemiological analysis of 228,344 knee injuries over 20 years","volume":"21","author":"Maniar","year":"2022","journal-title":"Lancet Reg Health West Pac"},{"key":"ref27","year":"2019"},{"key":"ref28","first-page":"2513","author":"Nurgazin","year":"2023"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1837","DOI":"10.1109\/TMI.2022.3150682","article-title":"Self-supervised learning for few-shot medical image segmentation","volume":"41","author":"Ouyang","year":"2022","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"61866","DOI":"10.1109\/ACCESS.2023.3283930","article-title":"Enhancing breast cancer classification in histopathological images through federated learning framework","volume":"11","author":"Peta","year":"2023","journal-title":"IEEE Access"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"1454","DOI":"10.1136\/bjsports-2018-100022","article-title":"Knee osteoarthritis risk is increased 4-6 fold after knee injury\u2013a systematic review and meta-analysis","volume":"53","author":"Poulsen","year":"2019","journal-title":"Br. J. Sports Med."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"4672","DOI":"10.1109\/JBHI.2023.3274498","article-title":"Federated semi supervised learning for medical image segmentation via pseudo-label denoising","volume":"27","author":"Qiu","year":"2023","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41746-020-00323-1","article-title":"The future of digital health with federated learning","volume":"3","author":"Rieke","year":"2020","journal-title":"NPJ Digit. Med."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"108111","DOI":"10.1016\/j.patcog.2021.108111","article-title":"MetaMed: few-shot medical image classification using gradient-based meta-learning","volume":"120","author":"Singh","year":"2021","journal-title":"Pattern Recogn."},{"key":"ref35","first-page":"1","article-title":"Prototypical networks for few-shot learning","volume":"30","author":"Snell","year":"2017","journal-title":"Adv. Neural Inf. Proces. Syst."},{"key":"ref36","first-page":"784","article-title":"Knee injury detection using MRI with efficiently-layered network (ELNet)","volume-title":"Medi. Imaging Deep Learn.","author":"Tsai","year":"2020"},{"key":"ref37","doi-asserted-by":"crossref","first-page":"2068","DOI":"10.1109\/TMI.2022.3227563","article-title":"Federated active learning for multicenter collaborative disease diagnosis","volume":"42","author":"Wu","year":"2022","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"1932","DOI":"10.1109\/TMI.2022.3233574","article-title":"Label-efficient self-supervised federated learning for tackling data heterogeneity in medical imaging","volume":"42","author":"Yan","year":"2023","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref39","first-page":"19","article-title":"Enhancing classification performance in knee magnetic resonance imaging using adversarial data augmentation","author":"Yan","year":"2023"},{"key":"ref40","first-page":"7419","article-title":"Gradient matching federated domain adaptation for brain image classification","volume":"35","author":"Zeng","year":"2022","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1589358\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,23]],"date-time":"2025-07-23T14:04:27Z","timestamp":1753279467000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1589358\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,23]]},"references-count":40,"alternative-id":["10.3389\/frai.2025.1589358"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1589358","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,23]]},"article-number":"1589358"}}