{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T20:17:18Z","timestamp":1793823438606,"version":"4.6.1"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2024,3,11]],"date-time":"2024-03-11T00:00:00Z","timestamp":1710115200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000923","name":"Australian Research Council","doi-asserted-by":"publisher","award":["IC190100020"],"award-info":[{"award-number":["IC190100020"]}],"id":[{"id":"10.13039\/501100000923","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:p>Musculoskeletal conditions affect an estimated 1.7 billion people worldwide, causing intense pain and disability. These conditions lead to 30 million emergency room visits yearly, and the numbers are only increasing. However, diagnosing musculoskeletal issues can be challenging, especially in emergencies where quick decisions are necessary. Deep learning (DL) has shown promise in various medical applications. However, previous methods had poor performance and a lack of transparency in detecting shoulder abnormalities on X-ray images due to a lack of training data and better representation of features. This often resulted in overfitting, poor generalisation, and potential bias in decision-making. To address these issues, a new trustworthy DL framework has been proposed to detect shoulder abnormalities (such as fractures, deformities, and arthritis) using X-ray images. The framework consists of two parts: same-domain transfer learning (TL) to mitigate imageNet mismatch and feature fusion to reduce error rates and improve trust in the final result. Same-domain TL involves training pre-trained models on a large number of labelled X-ray images from various body parts and fine-tuning them on the target dataset of shoulder X-ray images. Feature fusion combines the extracted features with seven DL models to train several ML classifiers. The proposed framework achieved an excellent accuracy rate of 99.2%, F1<jats:sub><jats:italic>Score<\/jats:italic><\/jats:sub> of 99.2%, and Cohen\u2019s kappa of 98.5%. Furthermore, the accuracy of the results was validated using three visualisation tools, including gradient-based class activation heat map (Grad CAM), activation visualisation, and locally interpretable model-independent explanations (LIME). The proposed framework outperformed previous DL methods and three orthopaedic surgeons invited to classify the test set, who obtained an average accuracy of 79.1%. The proposed framework has proven effective and robust, improving generalisation and increasing trust in the final results.<\/jats:p>","DOI":"10.1371\/journal.pone.0299545","type":"journal-article","created":{"date-parts":[[2024,3,11]],"date-time":"2024-03-11T17:27:41Z","timestamp":1710178061000},"page":"e0299545","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":35,"title":["Trustworthy deep learning framework for the detection of abnormalities in X-ray shoulder images"],"prefix":"10.1371","volume":"19","author":[{"given":"Laith","family":"Alzubaidi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Asma","family":"Salhi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9877-049X","authenticated-orcid":true,"given":"Mohammed","family":"A.Fadhel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinshuai","family":"Bai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Freek","family":"Hollman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kristine","family":"Italia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roberto","family":"Pareyon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"A. S.","family":"Albahri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7098-5480","authenticated-orcid":true,"given":"Chun","family":"Ouyang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jose","family":"Santamar\u00eda","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9784-8574","authenticated-orcid":true,"given":"Kenneth","family":"Cutbush","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ashish","family":"Gupta","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amin","family":"Abbosh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2770-5014","authenticated-orcid":true,"given":"Yuantong","family":"Gu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2024,3,11]]},"reference":[{"key":"pone.0299545.ref001","doi-asserted-by":"crossref","first-page":"106304","DOI":"10.1016\/j.cmpb.2021.106304","article-title":"Machine learning-based natural language processing of radiology reports in orthopaedic trauma","volume":"208","author":"AW Olthof","year":"2021","journal-title":"Computer methods and programs in biomedicine"},{"key":"pone.0299545.ref002","doi-asserted-by":"crossref","first-page":"106498","DOI":"10.1016\/j.cmpb.2021.106498","article-title":"Medical expert and machine learning analysis of lumbar disc herniation based on magnetic resonance imaging","volume":"213","author":"M Han","year":"2022","journal-title":"Computer Methods and Programs in Biomedicine"},{"key":"pone.0299545.ref003","first-page":"135","volume-title":"Rockwood and Matsen\u2019s The Shoulder","author":"K.L. Jensen","year":"2016"},{"key":"pone.0299545.ref004","unstructured":"Rajpurkar P, Irvin J, Bagul A, Ding D, Duan T, Mehta H, et al. Mura: Large dataset for abnormality detection in musculoskeletal radiographs. arXiv preprint arXiv:1712.06957. 2017 Dec 11. arXiv preprint arXiv:1712.06957. Available from: https:\/\/arxiv.org\/abs\/1712.06957"},{"key":"pone.0299545.ref005","unstructured":"WHO, World Health Organization:Musculoskeletal health, accessed: 01.10.2023. Available from: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/musculoskeletal-conditions"},{"issue":"45","key":"pone.0299545.ref006","doi-asserted-by":"crossref","first-page":"11591","DOI":"10.1073\/pnas.1806905115","article-title":"Deep neural network improves fracture detection by clinicians","volume":"115","author":"R Lindsey","year":"2018","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"3","key":"pone.0299545.ref007","doi-asserted-by":"crossref","first-page":"580","DOI":"10.3390\/healthcare10030580","article-title":"A novel hybrid machine learning based system to classify shoulder implant manufacturers","volume":"10","author":"E Sivari","year":"2022","journal-title":"Healthcare"},{"issue":"2","key":"pone.0299545.ref008","doi-asserted-by":"crossref","first-page":"681","DOI":"10.3390\/app12020681","article-title":"Deep learning for orthopedic disease based on medical image analysis: Present and future","volume":"12","author":"J Lee","year":"2022","journal-title":"Applied Sciences"},{"issue":"2","key":"pone.0299545.ref009","doi-asserted-by":"crossref","first-page":"215","DOI":"10.1080\/17453674.2019.1711323","article-title":"Deep learning in fracture detection: a narrative review","volume":"91","author":"PH Kalmet","year":"2020","journal-title":"Acta orthopaedica"},{"key":"pone.0299545.ref010","doi-asserted-by":"crossref","first-page":"106325","DOI":"10.1016\/j.cmpb.2021.106325","article-title":"In-depth learning of automatic segmentation of shoulder joint magnetic resonance images based on convolutional neural networks","volume":"211","author":"X Mu","year":"2021","journal-title":"Computer Methods and Programs in Biomedicine"},{"issue":"1","key":"pone.0299545.ref011","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1186\/s40537-023-00727-2","article-title":"A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications","volume":"10","author":"L Alzubaidi","year":"2023","journal-title":"Journal of Big Data"},{"key":"pone.0299545.ref012","first-page":"1","article-title":"Real-time diabetic foot ulcer classification based on deep learning & parallel hardware computational tools","author":"MA Fadhel","year":"2024","journal-title":"Multimedia Tools and Applications"},{"key":"pone.0299545.ref013","doi-asserted-by":"crossref","first-page":"107408","DOI":"10.1016\/j.cmpb.2023.107408","article-title":"Skin lesion segmentation using two-phase cross-domain transfer learning framework","volume":"231","author":"M Karri","year":"2023","journal-title":"Computer Methods and Programs in Biomedicine"},{"key":"pone.0299545.ref014","doi-asserted-by":"crossref","first-page":"106539","DOI":"10.1016\/j.cmpb.2021.106539","article-title":"Domain-and task-specific transfer learning for medical segmentation tasks","volume":"214","author":"R Zoetmulder","year":"2022","journal-title":"Computer Methods and Programs in Biomedicine"},{"issue":"5","key":"pone.0299545.ref015","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1016\/j.crad.2017.11.015","article-title":"Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks","volume":"73","author":"DH Kim","year":"2018","journal-title":"Clinical radiology"},{"issue":"13","key":"pone.0299545.ref016","doi-asserted-by":"crossref","first-page":"4523","DOI":"10.3390\/app10134523","article-title":"Towards a better understanding of transfer learning for medical imaging: a case study","volume":"10","author":"L Alzubaidi","year":"2020","journal-title":"Applied Sciences"},{"key":"pone.0299545.ref017","doi-asserted-by":"crossref","first-page":"e715","DOI":"10.7717\/peerj-cs.715","article-title":"Deepening into the suitability of using pre-trained models of ImageNet against a lightweight convolutional neural network in medical imaging: An experimental study","volume":"7","author":"L Alzubaidi","year":"2021","journal-title":"PeerJ Computer Science"},{"issue":"2","key":"pone.0299545.ref018","doi-asserted-by":"crossref","first-page":"570","DOI":"10.3390\/s23020570","article-title":"Incorporating a Novel Dual Transfer Learning Approach for Medical Images","volume":"23","author":"AA Mukhlif","year":"2023","journal-title":"Sensors"},{"key":"pone.0299545.ref019","article-title":"Transfusion: Understanding transfer learning for medical imaging","volume":"32","author":"M Raghu","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"pone.0299545.ref020","doi-asserted-by":"crossref","unstructured":"Azizi S, Mustafa B, Ryan F, Beaver Z, Freyberg J, Deaton J, et al. Big self-supervised models advance medical image classification. InProceedings of the IEEE\/CVF international conference on computer vision 2021 (pp. 3478\u20133488). Available from: https:\/\/ieeexplore.ieee.org\/abstract\/document\/9710396","DOI":"10.1109\/ICCV48922.2021.00346"},{"key":"pone.0299545.ref021","first-page":"1","article-title":"Robust and data-efficient generalization of self-supervised machine learning for diagnostic imaging","author":"S Azizi","year":"2023","journal-title":"Nature Biomedical Engineering"},{"key":"pone.0299545.ref022","doi-asserted-by":"crossref","first-page":"58667","DOI":"10.1109\/ACCESS.2021.3070701","article-title":"EGFR assessment in lung cancer CT images: analysis of local and holistic regions of interest using deep unsupervised transfer learning","volume":"9","author":"F Silva","year":"2021","journal-title":"IEEE Access"},{"issue":"11","key":"pone.0299545.ref023","doi-asserted-by":"crossref","first-page":"1908","DOI":"10.1093\/jamia\/ocac135","article-title":"Transfer learning enables prediction of myocardial injury from continuous single-lead electrocardiography","volume":"29","author":"BT Jin","year":"2022","journal-title":"Journal of the American Medical Informatics Association"},{"issue":"7","key":"pone.0299545.ref024","doi-asserted-by":"crossref","first-page":"1590","DOI":"10.3390\/cancers13071590","article-title":"Novel transfer learning approach for medical imaging with limited labeled data","volume":"13","author":"L Alzubaidi","year":"2021","journal-title":"Cancers"},{"issue":"12","key":"pone.0299545.ref025","doi-asserted-by":"crossref","first-page":"1399","DOI":"10.1038\/s41551-022-00936-9","article-title":"Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning","volume":"6","author":"E Tiu","year":"2022","journal-title":"Nature Biomedical Engineering"},{"key":"pone.0299545.ref026","doi-asserted-by":"crossref","first-page":"102256","DOI":"10.1016\/j.media.2021.102256","article-title":"Self-supervised driven consistency training for annotation efficient histopathology image analysis","volume":"75","author":"CL Srinidhi","year":"2022","journal-title":"Medical Image Analysis"},{"issue":"12","key":"pone.0299545.ref027","doi-asserted-by":"crossref","first-page":"1346","DOI":"10.1038\/s41551-022-00914-1","article-title":"Self-supervised learning in medicine and healthcare","volume":"6","author":"R Krishnan","year":"2022","journal-title":"Nature Biomedical Engineering"},{"key":"pone.0299545.ref028","first-page":"122807","article-title":"A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations","author":"Z Zhao","year":"2023","journal-title":"Expert Systems with Applications"},{"issue":"1","key":"pone.0299545.ref029","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1186\/s12880-022-00793-7","article-title":"Transfer learning for medical image classification: a literature review","volume":"22","author":"HE Kim","year":"2022","journal-title":"BMC medical imaging"},{"issue":"6","key":"pone.0299545.ref030","doi-asserted-by":"crossref","first-page":"2723","DOI":"10.3390\/app11062723","article-title":"Classification of shoulder x-ray images with deep learning ensemble models","volume":"11","author":"F Uysal","year":"2021","journal-title":"Applied Sciences"},{"issue":"11","key":"pone.0299545.ref031","doi-asserted-by":"crossref","first-page":"127","DOI":"10.3390\/jimaging6110127","article-title":"Musculoskeletal images classification for detection of fractures using transfer learning","volume":"6","author":"I Kandel","year":"2020","journal-title":"Journal of imaging"},{"issue":"6","key":"pone.0299545.ref032","doi-asserted-by":"crossref","first-page":"100","DOI":"10.3390\/jimaging7060100","article-title":"Comparing stacking ensemble techniques to improve musculoskeletal fracture image classification","volume":"7","author":"I Kandel","year":"2021","journal-title":"Journal of Imaging"},{"issue":"9","key":"pone.0299545.ref033","doi-asserted-by":"crossref","first-page":"175","DOI":"10.3390\/computers12090175","article-title":"Novel deep feature fusion framework for multi-scenario violence detection","volume":"12","author":"SA Jebur","year":"2023","journal-title":"Computers"},{"key":"pone.0299545.ref034","doi-asserted-by":"crossref","first-page":"107627","DOI":"10.1016\/j.engappai.2023.107627","article-title":"Reliable deep learning framework for the ground penetrating radar data to locate the horizontal variation in levee soil compaction","volume":"129","author":"L Alzubaidi","year":"2024","journal-title":"Engineering Applications of Artificial Intelligence"},{"key":"pone.0299545.ref035","doi-asserted-by":"crossref","unstructured":"Alzubaidi L, Fadhel MA, Albahri AS, Salhi A, Gupta A, Gu Y. Domain Adaptation and Feature Fusion for the Detection of Abnormalities in X-Ray Forearm Images. In 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) 2023 Jul 24 (pp. 1\u20135). IEEE. Available from: https:\/\/doi.org\/10.1109\/EMBC40787.2023.10340309.","DOI":"10.1109\/EMBC40787.2023.10340309"},{"issue":"10","key":"pone.0299545.ref036","doi-asserted-by":"crossref","first-page":"1689","DOI":"10.3390\/diagnostics13101689","article-title":"A Deep Feature Fusion of Improved Suspected Keratoconus Detection with Deep Learning","volume":"13","author":"AH Al-Timemy","year":"2023","journal-title":"Diagnostics"},{"key":"pone.0299545.ref037","doi-asserted-by":"crossref","DOI":"10.1155\/2023\/4459198","article-title":"Towards risk-free trustworthy artificial intelligence: Significance and requirements","volume":"2023","author":"L Alzubaidi","year":"2023","journal-title":"International Journal of Intelligent Systems"},{"key":"pone.0299545.ref038","doi-asserted-by":"crossref","first-page":"107161","DOI":"10.1016\/j.cmpb.2022.107161","article-title":"Application of explainable artificial intelligence for healthcare: A systematic review of the last decade (2011\u20132022)","author":"HW Loh","year":"2022","journal-title":"Computer Methods and Programs in Biomedicine"},{"issue":"8","key":"pone.0299545.ref039","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1038\/s42256-022-00516-1","article-title":"Advances, challenges and opportunities in creating data for trustworthy AI","volume":"4","author":"W Liang","year":"2022","journal-title":"Nature Machine Intelligence"},{"issue":"10","key":"pone.0299545.ref040","doi-asserted-by":"crossref","first-page":"867","DOI":"10.1038\/s42256-022-00536-x","article-title":"Benchmarking saliency methods for chest X-ray interpretation","volume":"4","author":"A Saporta","year":"2022","journal-title":"Nature Machine Intelligence"},{"key":"pone.0299545.ref041","article-title":"A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion","author":"AS Albahri","year":"2023","journal-title":"Information Fusion"},{"issue":"12","key":"pone.0299545.ref042","doi-asserted-by":"crossref","first-page":"578","DOI":"10.1038\/s42256-019-0126-0","article-title":"Automated abnormality detection in lower extremity radiographs using deep learning","volume":"1","author":"M Varma","year":"2019","journal-title":"Nature Machine Intelligence"},{"issue":"1","key":"pone.0299545.ref043","doi-asserted-by":"crossref","first-page":"9097","DOI":"10.1038\/s41598-021-88578-w","article-title":"A calibrated deep learning ensemble for abnormality detection in musculoskeletal radiographs","volume":"11","author":"M He","year":"2021","journal-title":"Scientific Reports"},{"issue":"18","key":"pone.0299545.ref044","doi-asserted-by":"crossref","first-page":"3291","DOI":"10.3390\/math10183291","article-title":"Fractured elbow classification using hand-crafted and deep feature fusion and selection based on whale optimization approach","volume":"10","author":"S Malik","year":"2022","journal-title":"Mathematics"},{"key":"pone.0299545.ref045","doi-asserted-by":"crossref","unstructured":"Manoila CP, Ciurea A, Albu F. SmartMRI Framework for Segmentation of MR Images Using Multiple Deep Learning Methods. In2022 E-Health and Bioengineering Conference (EHB) 2022 Nov 17 (pp. 01\u201304). IEEE. Available from: https:\/\/doi.org\/10.1109\/EHB55594.2022.9991496","DOI":"10.1109\/EHB55594.2022.9991496"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0299545","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,11]],"date-time":"2024-03-11T17:28:40Z","timestamp":1710178120000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0299545"}},"subtitle":[],"editor":[{"given":"Ali Mohammad","family":"Alqudah","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,3,11]]},"references-count":45,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2024,3,11]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0299545","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,3,11]]}}}