{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,27]],"date-time":"2026-06-27T09:26:37Z","timestamp":1782552397226,"version":"3.54.5"},"reference-count":21,"publisher":"Public Library of Science (PLoS)","issue":"8","license":[{"start":{"date-parts":[[2023,8,30]],"date-time":"2023-08-30T00:00:00Z","timestamp":1693353600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:p>In this study, we present a deep learning model for fracture classification on shoulder radiographs using a convolutional neural network (CNN). The primary aim was to evaluate the classification performance of the CNN for proximal humeral fractures (PHF) based on the AO\/OTA classification system. Secondary objectives included evaluating the model\u2019s performance for diaphyseal humerus, clavicle, and scapula fractures. The training dataset consisted of 6,172 examinations, including 2\u20137 radiographs per examination. The overall area under the curve (AUC) for fracture classification was 0.89, indicating good performance. For PHF classification, 12 out of 16 classes achieved an AUC of 0.90 or greater. Additionally, the CNN model had excellent overall AUC for diaphyseal humerus fractures (0.97), clavicle fractures (0.96), and good AUC for scapula fractures (0.87). Despite the limitations of the study, such as the reliance on ground truth labels provided by students with limited radiographic assessment experience, our findings are in concordance with previous studies, further consolidating CNN as potent fracture classifiers in plain radiographs. The inclusion of multiple radiographs with different views from each examination, as well as the generally unselected nature of the sample, contributed to the overall generalizability of the study. This is the fifth study published by our group on AI in orthopaedic radiographs, which has consistently shown promising results. The next challenge for the orthopaedic research community will be to transfer these results from the research setting into clinical practice. External validation of the CNN model should be conducted in the future before it is considered for use in a clinical setting.<\/jats:p>","DOI":"10.1371\/journal.pone.0289808","type":"journal-article","created":{"date-parts":[[2023,8,30]],"date-time":"2023-08-30T17:33:11Z","timestamp":1693416791000},"page":"e0289808","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":18,"title":["Deep learning classification of shoulder fractures on plain radiographs of the humerus, scapula and clavicle"],"prefix":"10.1371","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0341-0227","authenticated-orcid":true,"given":"Martin","family":"Magn\u00e9li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Petter","family":"Ling","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jacob","family":"Gisl\u00e9n","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Johan","family":"Fagrell","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yilmaz","family":"Demir","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erica Domeij","family":"Arverud","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-6617-1963","authenticated-orcid":true,"given":"Kristofer","family":"Hallberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3437-7290","authenticated-orcid":true,"given":"Bj\u00f6rn","family":"Salomonsson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8080-5815","authenticated-orcid":true,"given":"Max","family":"Gordon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2023,8,30]]},"reference":[{"key":"pone.0289808.ref001","article-title":"Proximal humeral fracture osteosynthesis in Belgium: a retrospective population-based epidemiologic study.","author":"J Dauwe","year":"2020","journal-title":"Eur J Trauma Emerg Surg"},{"key":"pone.0289808.ref002","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1007\/BF00932316","article-title":"The epidemiology of fractures of the proximal humerus.","volume":"108","author":"T Lind","year":"1989","journal-title":"Arch Orthop Trauma Surg"},{"key":"pone.0289808.ref003","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1186\/s12891-015-0564-8","article-title":"A scoping review of the proximal humerus fracture literature.","volume":"16","author":"GP Slobogean","year":"2015","journal-title":"BMC Musculoskelet Disord"},{"key":"pone.0289808.ref004","doi-asserted-by":"crossref","first-page":"S1","DOI":"10.1097\/BOT.0000000000001063","article-title":"Fracture and Dislocation Classification Compendium-2018.","volume":"321","author":"EG Meinberg","year":"2018","journal-title":"J Orthop Trauma."},{"key":"pone.0289808.ref005","doi-asserted-by":"crossref","first-page":"1077","DOI":"10.2106\/00004623-197052060-00001","article-title":"Displaced proximal humeral fractures. I. Classification and evaluation","volume":"52","author":"CS Neer","year":"1970","journal-title":"J Bone Joint Surg Am"},{"key":"pone.0289808.ref006","doi-asserted-by":"crossref","first-page":"1090","DOI":"10.2106\/00004623-197052060-00002","article-title":"Displaced proximal humeral fractures. II. Treatment of three-part and four-part displacement","volume":"52","author":"CS Neer","year":"1970","journal-title":"J Bone Joint Surg Am"},{"key":"pone.0289808.ref007","doi-asserted-by":"crossref","first-page":"393","DOI":"10.1016\/j.ocl.2008.05.002","article-title":"Classification and imaging of proximal humerus fractures.","volume":"39","author":"BC Robinson","year":"2008","journal-title":"Orthop Clin North Am"},{"key":"pone.0289808.ref008","doi-asserted-by":"crossref","first-page":"1751","DOI":"10.2106\/00004623-199312000-00003","article-title":"The reproducibility of classification of fractures of the proximal end of the humerus","volume":"75","author":"KA Siebenrock","year":"1993","journal-title":"J Bone Joint Surg Am"},{"key":"pone.0289808.ref009","doi-asserted-by":"crossref","first-page":"239","DOI":"10.3109\/17453679708996692","article-title":"Poor reproducibility of classification of proximal humeral fractures. Additional CT of minor value","volume":"68","author":"GO Sj\u00f6d\u00e9n","year":"1997","journal-title":"Acta Orthop Scand."},{"key":"pone.0289808.ref010","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1186\/s10195-020-0543-1","article-title":"Reliability and reproducibility of the new AO\/OTA 2018 classification system for proximal humeral fractures: a comparison of three different classification systems.","volume":"21","author":"G Marongiu","year":"2020","journal-title":"J Orthop Traumatol"},{"key":"pone.0289808.ref011","doi-asserted-by":"crossref","first-page":"452","DOI":"10.1067\/mse.2002.126613","article-title":"Epidemiology of clavicle fractures.","volume":"11","author":"F Postacchini","year":"2002","journal-title":"J Shoulder Elbow Surg"},{"key":"pone.0289808.ref012","first-page":"127","article-title":"The incidence of fractures of the clavicle","author":"A Nordqvist","year":"1994","journal-title":"Clin Orthop Relat Res"},{"key":"pone.0289808.ref013","doi-asserted-by":"crossref","first-page":"e1002707","DOI":"10.1371\/journal.pmed.1002707","article-title":"Deep learning and artificial intelligence in radiology: Current applications and future directions","volume":"15","author":"K Yasaka","year":"2018","journal-title":"PLoS Med"},{"key":"pone.0289808.ref014","doi-asserted-by":"crossref","first-page":"468","DOI":"10.1080\/17453674.2018.1453714","article-title":"Automated detection and classification of the proximal humerus fracture by using deep learning algorithm.","volume":"89","author":"SW Chung","year":"2018","journal-title":"Acta Orthopaedica"},{"key":"pone.0289808.ref015","doi-asserted-by":"crossref","first-page":"1026","DOI":"10.1109\/ICCV.2015.123","article-title":"Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification.","author":"K He","year":"2015","journal-title":"2015 IEEE International Conference on Computer Vision (ICCV)."},{"key":"pone.0289808.ref016","doi-asserted-by":"crossref","first-page":"1315","DOI":"10.1097\/JTO.0b013e3181ec173d","article-title":"Receiver Operating Characteristic Curve in Diagnostic Test Assessment","volume":"5","author":"JN Mandrekar","year":"2010","journal-title":"Journal of Thoracic Oncology"},{"key":"pone.0289808.ref017","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1177\/001316446002000104","article-title":"A Coefficient of Agreement for Nominal Scales","volume":"20","author":"J. Cohen","year":"1960","journal-title":"Educational and Psychological Measurement"},{"key":"pone.0289808.ref018","doi-asserted-by":"crossref","first-page":"581","DOI":"10.1080\/17453674.2017.1344459","article-title":"Artificial intelligence for analyzing orthopedic trauma radiographs","volume":"88","author":"J Olczak","year":"2017","journal-title":"Acta Orthopaedica"},{"key":"pone.0289808.ref019","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1080\/17453674.2020.1837420","article-title":"Ankle fracture classification using deep learning: automating detailed AO Foundation\/Orthopedic Trauma Association (AO\/OTA) 2018 malleolar fracture identification reaches a high degree of correct classification.","volume":"92","author":"J Olczak","year":"2021","journal-title":"Acta Orthop"},{"key":"pone.0289808.ref020","doi-asserted-by":"crossref","first-page":"2482","DOI":"10.1097\/CORR.0000000000000848","article-title":"What Are the Applications and Limitations of Artificial Intelligence for Fracture Detection and Classification in Orthopaedic Trauma Imaging? A Systematic Review","volume":"477","author":"DWG Langerhuizen","year":"2019","journal-title":"Clin Orthop Relat Res"},{"key":"pone.0289808.ref021","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1007\/s00256-018-3016-3","article-title":"Detecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network.","volume":"48","author":"T Urakawa","year":"2019","journal-title":"Skeletal Radiol"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0289808","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,30]],"date-time":"2023-08-30T17:33:26Z","timestamp":1693416806000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0289808"}},"subtitle":[],"editor":[{"given":"Hans-Peter","family":"Simmen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2023,8,30]]},"references-count":21,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2023,8,30]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0289808","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,30]]}}}