{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T09:00:11Z","timestamp":1781168411709,"version":"3.54.1"},"reference-count":33,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T00:00:00Z","timestamp":1781136000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Zhejiang Provincial Natural Science Foundation of China","award":["LQ22H100003"],"award-info":[{"award-number":["LQ22H100003"]}]},{"name":"Zhejiang Provincial Health Commission","award":["2020KY132"],"award-info":[{"award-number":["2020KY132"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Spinal infections are rare but serious conditions requiring timely diagnosis. Non-contrast computed tomography (CT) is widely used and may incidentally reveal spinal abnormalities; however, subtle infectious findings are often missed, especially on axial images without sagittal reconstruction.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>To investigate a deep learning approach for diagnosing primary spinal infections using non-contrast CT images.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      This retrospective dual-center study included 157 patients with primary spinal infection. A Swin Transformer model was developed using non-contrast CT slices. Patients from the primary center (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a0127) were randomly split 7:3 into training and internal validation sets. An independent external cohort (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a030) from a second center was used for external validation. Per-slice diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC), area under the precision-recall curve (AUPRC), sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and F1-score, with a probability threshold \u22650.5 determined by the Youden index. Model performance was compared with that of two radiologists.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The Swin Transformer model demonstrated excellent per-slice diagnostic performance. In the internal validation set, the model achieved an AUC of 0.979, sensitivity of 96.2%, specificity of 90.5%, and accuracy of 89.4%. In the external cohort, similar results were obtained: AUC 0.989, sensitivity 98.2%, specificity 98.3%, and accuracy 98.3%. The deep learning model significantly outperformed both radiologists in AUC and sensitivity across cohorts (all\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a00.05). With artificial intelligence (AI) assistance, both radiologists showed substantial improvements in diagnostic performance and efficiency in both internal and external validation (all\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a00.05). The model\u2019s reading time was markedly shorter than that of unassisted radiologists, and AI assistance reduced radiologists\u2019 reading time (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a00.001). Presence of spinal epidural abscess and pathogen type significantly influenced diagnostic outcomes (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a00.001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>This Swin Transformer-based deep learning model achieves high diagnostic accuracy for detecting spinal infections on axial non-contrast CT images, with performance comparable to or exceeding that of musculoskeletal radiologists. By enhancing radiologists\u2019 sensitivity and reducing reading time, the model shows promise as a clinical decision support tool to reduce missed diagnoses, particularly in emergency or resource-limited settings where magnetic resonance imaging (MRI) is unavailable. Its robust performance on external validation supports generalizability and lays the foundation for future multicenter prospective studies and extension to other spinal pathologies.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.7717\/peerj.21340","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T08:33:47Z","timestamp":1781166827000},"page":"e21340","source":"Crossref","is-referenced-by-count":0,"title":["A transformer-based deep learning algorithm for diagnosing spinal infections on axial non-contrast computed tomography images: a dual-center retrospective study"],"prefix":"10.7717","volume":"14","author":[{"given":"Dongdong","family":"Yu","sequence":"first","affiliation":[{"name":"Department of Orthopedic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine,  Hangzhou, Zhe Jiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenting","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Stevens Institute of Technology, Hoboken, New Jersey, United States of America"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kai","family":"Song","sequence":"additional","affiliation":[{"name":"The First Affiliated Hospital of Henan Medical University, Xin Xiang, He Nan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenxin","family":"Lu","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine,  Hangzhou, Zhe Jiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingfeng","family":"Xu","sequence":"additional","affiliation":[{"name":"Department of Radiology, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou,  Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiali","family":"Zheng","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine,  Hangzhou, Zhe Jiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongjian","family":"Kang","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine,  Hangzhou, Zhe Jiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuan","family":"Hong","sequence":"additional","affiliation":[{"name":"College of Mathematical Medicine, Zhejiang Normal University, Jinhua, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bin","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, First Affiliated Hospital, Zhejiang University School of Medicine,  Hangzhou, Zhe Jiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,6,11]]},"reference":[{"key":"10.7717\/peerj.21340\/ref-1","doi-asserted-by":"publisher","first-page":"508","DOI":"10.1016\/j.jor.2019.08.014","article-title":"Management of pyogenic spinal infection, review of literature","volume":"16","author":"Aljawadi","year":"2019","journal-title":"Journal of Orthopaedics"},{"key":"10.7717\/peerj.21340\/ref-2","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1016\/j.idc.2017.01.003","article-title":"Infections of the spine","volume":"31","author":"Babic","year":"2017","journal-title":"Infectious Disease Clinics of North America"},{"key":"10.7717\/peerj.21340\/ref-3","doi-asserted-by":"publisher","first-page":"265","DOI":"10.1097\/QCO.0000000000000544","article-title":"Update on spinal epidural abscess","volume":"32","author":"Babic","year":"2019","journal-title":"Current Opinion in Infectious Diseases"},{"key":"10.7717\/peerj.21340\/ref-4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.12968\/hmed.2021.0448","article-title":"Spondylodiscitis in adults: diagnosis and management","volume":"83","author":"Baryeh","year":"2022","journal-title":"British Journal of Hospital Medicine"},{"key":"10.7717\/peerj.21340\/ref-5","doi-asserted-by":"publisher","first-page":"e26","DOI":"10.1093\/cid\/civ482","article-title":"2015 Infectious Diseases Society of America (IDSA) clinical practice guidelines for the diagnosis and treatment of native vertebral osteomyelitis in adults","volume":"61","author":"Berbari","year":"2015","journal-title":"Clinical Infectious Diseases"},{"key":"10.7717\/peerj.21340\/ref-6","doi-asserted-by":"publisher","first-page":"975","DOI":"10.1016\/j.amjmed.2017.03.009","article-title":"Errors in diagnosis of spinal epidural abscesses in the era of electronic health records","volume":"130","author":"Bhise","year":"2017","journal-title":"The American Journal of Medicine"},{"key":"10.7717\/peerj.21340\/ref-7","doi-asserted-by":"publisher","first-page":"120","DOI":"10.1148\/radiol.13121632","article-title":"Unreported vertebral body compression fractures at abdominal multidetector CT","volume":"268","author":"Carberry","year":"2013","journal-title":"Radiology"},{"key":"10.7717\/peerj.21340\/ref-8","doi-asserted-by":"publisher","first-page":"226","DOI":"10.14444\/7019","article-title":"Spondylodiscitis: a retrospective analysis of clinical, etiological, and radiological diagnosis","volume":"14","author":"Cardoso","year":"2020","journal-title":"International Journal of Spine Surgery"},{"key":"10.7717\/peerj.21340\/ref-9","doi-asserted-by":"publisher","first-page":"893","DOI":"10.3390\/microorganisms12050893","article-title":"Imaging of spondylodiscitis: a comprehensive updated review-multimodality imaging findings, differential diagnosis, and specific microorganisms detection","volume":"12","author":"Cromb\u00e9","year":"2024","journal-title":"Microorganisms"},{"key":"10.7717\/peerj.21340\/ref-10","doi-asserted-by":"publisher","first-page":"2000045","DOI":"10.1183\/13993003.00045-2020","article-title":"Identification of potentially undiagnosed patients with nontuberculous mycobacterial lung disease using machine learning applied to primary care data in the UK","volume":"56","author":"Doyle","year":"2020","journal-title":"The European Respiratory Journal"},{"key":"10.7717\/peerj.21340\/ref-11","doi-asserted-by":"publisher","first-page":"5355","DOI":"10.1109\/TNNLS.2022.3204090","article-title":"SwinPA-Net: swin transformer-based multiscale feature pyramid aggregation network for medical image segmentation","volume":"35","author":"Du","year":"2024","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"10.7717\/peerj.21340\/ref-12","doi-asserted-by":"publisher","first-page":"2787","DOI":"10.1007\/s00586-013-2850-1","article-title":"Spinal infection: state of the art and management algorithm","volume":"22","author":"Duarte","year":"2013","journal-title":"European Spine Journal"},{"key":"10.7717\/peerj.21340\/ref-13","doi-asserted-by":"publisher","first-page":"185","DOI":"10.4103\/jcvjs.jcvjs_164_23","article-title":"Spondylodiscitis: Understanding pathophysiology, surgical strategies, and postoperative management - a single-center study","volume":"15","author":"Encarnaci\u00f3n-Santos","year":"2024","journal-title":"Journal of Craniovertebral Junction & Spine"},{"key":"10.7717\/peerj.21340\/ref-14","doi-asserted-by":"publisher","first-page":"S488","DOI":"10.1016\/j.jacr.2021.09.001","article-title":"ACR appropriateness criteria? Suspected spine infection","volume":"18","author":"Expert","year":"2021","journal-title":"Journal of the American College of Radiology: JACR"},{"key":"10.7717\/peerj.21340\/ref-15","doi-asserted-by":"publisher","first-page":"1201","DOI":"10.1056\/NEJMra2302038","article-title":"Artificial intelligence and machine learning in clinical medicine. 2023","volume":"388","author":"Haug","year":"2023","journal-title":"The New England Journal of Medicine"},{"key":"10.7717\/peerj.21340\/ref-16","doi-asserted-by":"publisher","first-page":"13124","DOI":"10.1038\/s41598-018-31486-3","article-title":"Performance of the deep convolutional neural network based magnetic resonance image scoring algorithm for differentiating between tuberculous and pyogenic spondylitis","volume":"8","author":"Kim","year":"2018","journal-title":"Scientific Reports"},{"key":"10.7717\/peerj.21340\/ref-17","doi-asserted-by":"publisher","first-page":"284","DOI":"10.1038\/s41597-021-01060-0","article-title":"A computed tomography vertebral segmentation dataset with anatomical variations and multi-vendor scanner data","volume":"8","author":"Liebl","year":"2021","journal-title":"Scientific Data"},{"key":"10.7717\/peerj.21340\/ref-18","doi-asserted-by":"publisher","first-page":"31","DOI":"10.2147\/IDR.S482584","article-title":"Deep learning for discrimination of early spinal tuberculosis from acute osteoporotic vertebral fracture on CT","volume":"18","author":"Liu","year":"2025","journal-title":"Infection and Drug Resistance"},{"key":"10.7717\/peerj.21340\/ref-19","doi-asserted-by":"publisher","first-page":"2599329","DOI":"10.1080\/24699322.2025.2599329","article-title":"Deep learning diagnosis model of spinal tuberculosis based on CT bone window gradient attention mechanism: multi-center study","volume":"31","author":"Mo","year":"2026","journal-title":"Computer Assisted Surgery"},{"key":"10.7717\/peerj.21340\/ref-20","doi-asserted-by":"publisher","first-page":"390","DOI":"10.1097\/BRS.0000000000004889","article-title":"A new deep learning algorithm for detecting spinal metastases on computed tomography images","volume":"49","author":"Motohashi","year":"2024","journal-title":"Spine"},{"key":"10.7717\/peerj.21340\/ref-21","doi-asserted-by":"publisher","first-page":"1696","DOI":"10.1007\/s00330-008-0920-2","article-title":"Significance of sagittal reformations in routine thoracic and abdominal multislice CT studies for detecting osteoporotic fractures and other spine abnormalities","volume":"18","author":"M\u00fcller","year":"2008","journal-title":"European Radiology"},{"key":"10.7717\/peerj.21340\/ref-22","doi-asserted-by":"publisher","first-page":"2963","DOI":"10.3390\/diagnostics15232963","article-title":"Multimodal-imaging-based interpretable deep learning framework for distinguishing brucella from tuberculosis spondylitis: a dual-center study","volume":"15","author":"Nijiati","year":"2025","journal-title":"Diagnostics"},{"key":"10.7717\/peerj.21340\/ref-23","doi-asserted-by":"publisher","first-page":"618","DOI":"10.1186\/s12879-024-09462-2","article-title":"Combined clinical significance of MRI and serum mannose-binding lectin in the prediction of spinal tuberculosis","volume":"24","author":"Qi","year":"2024","journal-title":"BMC Infectious Diseases"},{"key":"10.7717\/peerj.21340\/ref-24","doi-asserted-by":"publisher","first-page":"3454","DOI":"10.21037\/qims-21-939","article-title":"Current development and prospects of deep learning in spine image analysis: a literature review","volume":"12","author":"Qu","year":"2022","journal-title":"Quantitative Imaging in Medicine and Surgery"},{"key":"10.7717\/peerj.21340\/ref-25","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1053\/j.semnuclmed.2022.11.005","article-title":"Imaging of spondylodiscitis: an update","volume":"53","author":"Raghavan","year":"2023","journal-title":"Seminars in Nuclear Medicine"},{"key":"10.7717\/peerj.21340\/ref-26","doi-asserted-by":"publisher","first-page":"843","DOI":"10.1007\/s11547-021-01347-7","article-title":"Differentiation between infectious spondylodiscitis versus inflammatory or degenerative spinal changes: how can magnetic resonance imaging help the clinician?","volume":"126","author":"Salaffi","year":"2021","journal-title":"La Radiologia Medica"},{"key":"10.7717\/peerj.21340\/ref-27","doi-asserted-by":"publisher","first-page":"e230088","DOI":"10.1148\/ryai.230088","article-title":"Vision transformer-based decision support for neurosurgical intervention in acute traumatic brain injury: automated surgical intervention support tool, radiology","volume":"6","author":"Smith","year":"2024","journal-title":"Artificial Intelligence"},{"key":"10.7717\/peerj.21340\/ref-28","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1186\/s41747-024-00470-0","article-title":"Deep learning reconstruction for lumbar spine MRI acceleration: a prospective study","volume":"8","author":"Tang","year":"2024","journal-title":"European Radiology Experimental"},{"key":"10.7717\/peerj.21340\/ref-29","doi-asserted-by":"publisher","first-page":"476","DOI":"10.3390\/microorganisms8040476","article-title":"Spinal infections: an update","volume":"8","author":"Tsantes","year":"2020","journal-title":"Microorganisms"},{"key":"10.7717\/peerj.21340\/ref-30","doi-asserted-by":"publisher","first-page":"4625","DOI":"10.3390\/jcm11154625","article-title":"Swin Transformer improves the IDH mutation status prediction of gliomas free of MRI-based tumor segmentation","volume":"11","author":"Wu","year":"2022","journal-title":"Journal of Clinical Medicine"},{"key":"10.7717\/peerj.21340\/ref-31","doi-asserted-by":"publisher","first-page":"3702","DOI":"10.1007\/s00586-025-08696-1","article-title":"Deep learning radiomics model based on contrast-enhanced MRI for distinguishing between tuberculous spondylitis and pyogenic spondylitis","volume":"34","author":"Yang","year":"2025","journal-title":"European Spine Journal"},{"key":"10.7717\/peerj.21340\/ref-32","doi-asserted-by":"publisher","first-page":"102772","DOI":"10.1016\/j.eclinm.2024.102772","article-title":"Development and validation of a deep learning-based framework for automated lung CT segmentation and acute respiratory distress syndrome prediction: a multicenter cohort study","volume":"75","author":"Zhou","year":"2024","journal-title":"EClinicalMedicine"},{"key":"10.7717\/peerj.21340\/ref-33","doi-asserted-by":"publisher","first-page":"107225","DOI":"10.1016\/j.ijid.2024.107225","article-title":"Melampsora spondylitis presenting with unexplained low back pain: a case report","volume":"148","author":"Zhu","year":"2024","journal-title":"International Journal of Infectious Diseases"}],"container-title":["PeerJ"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/21340.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21340.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21340.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21340.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T08:33:51Z","timestamp":1781166831000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/21340"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,11]]},"references-count":33,"alternative-id":["10.7717\/peerj.21340"],"URL":"https:\/\/doi.org\/10.7717\/peerj.21340","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2167-8359"],"issn-type":[{"value":"2167-8359","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,11]]},"article-number":"e21340"}}