{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,4]],"date-time":"2026-11-04T20:51:47Z","timestamp":1793825507386,"version":"4.6.1"},"reference-count":30,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"7","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":["Clin Orthop Relat Res"],"published-print":{"date-parts":[[2021,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Vertebral fractures are the most common osteoporotic fractures in older individuals. Recent studies suggest that the performance of artificial intelligence is equal to humans in detecting osteoporotic fractures, such as fractures of the hip, distal radius, and proximal humerus. However, whether artificial intelligence performs as well in the detection of vertebral fractures on plain lateral spine radiographs has not yet been reported.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Questions\/purposes<\/jats:title>\n                    <jats:p>(1) What is the accuracy, sensitivity, specificity, and interobserver reliability (kappa value) of an artificial intelligence model in detecting vertebral fractures, based on Genant fracture grades, using plain lateral spine radiographs compared with values obtained by human observers? (2) Do patients\u2019 clinical data, including the anatomic location of the fracture (thoracic or lumbar spine), T-score on dual-energy x-ray absorptiometry, or fracture grade severity, affect the performance of an artificial intelligence model? (3) How does the artificial intelligence model perform on external validation?<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Between 2016 and 2018, 1019 patients older than 60 years were treated for vertebral fractures in our institution. Seventy-eight patients were excluded because of missing CT or MRI scans (24% [19]), poor image quality in plain lateral radiographs of spines (54% [42]), multiple myeloma (5% [4]), and prior spine instrumentation (17% [13]). The plain lateral radiographs of 941 patients (one radiograph per person), with a mean age of 76 \u00b1 12 years, and 1101 vertebral fractures between T7 and L5 were retrospectively evaluated for training (n = 565), validating (n = 188), and testing (n = 188) of an artificial intelligence deep-learning model. The gold standard for diagnosis (ground truth) of a vertebral fracture is the interpretation of the CT or MRI reports by a spine surgeon and a radiologist independently. If there were any disagreements between human observers, the corresponding CT or MRI images would be rechecked by them together to reach a consensus. For the Genant classification, the injured vertebral body height was measured in the anterior, middle, and posterior third. Fractures were classified as Grade 1 (&lt; 25%), Grade 2 (26% to 40%), or Grade 3 (&gt; 40%). The framework of the artificial intelligence deep-learning model included object detection, data preprocessing of radiographs, and classification to detect vertebral fractures. Approximately 90 seconds was needed to complete the procedure and obtain the artificial intelligence model results when applied clinically. The accuracy, sensitivity, specificity, interobserver reliability (kappa value), receiver operating characteristic curve, and area under the curve (AUC) were analyzed. The bootstrapping method was applied to our testing dataset and external validation dataset. The accuracy, sensitivity, and specificity were used to investigate whether fracture anatomic location or T-score in dual-energy x-ray absorptiometry report affected the performance of the artificial intelligence model. The receiver operating characteristic curve and AUC were used to investigate the relationship between the performance of the artificial intelligence model and fracture grade. External validation with a similar age population and plain lateral radiographs from another medical institute was also performed to investigate the performance of the artificial intelligence model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The artificial intelligence model with ensemble method demonstrated excellent accuracy (93% [773 of 830] of vertebrae), sensitivity (91% [129 of 141]), and specificity (93% [644 of 689]) for detecting vertebral fractures of the lumbar spine. The interobserver reliability (kappa value) of the artificial intelligence performance and human observers for thoracic and lumbar vertebrae were 0.72 (95% CI 0.65 to 0.80; p &lt; 0.001) and 0.77 (95% CI 0.72 to 0.83; p &lt; 0.001), respectively. The AUCs for Grades 1, 2, and 3 vertebral fractures were 0.919, 0.989, and 0.990, respectively. The artificial intelligence model with ensemble method demonstrated poorer performance for discriminating normal osteoporotic lumbar vertebrae, with a specificity of 91% (260 of 285) compared with nonosteoporotic lumbar vertebrae, with a specificity of 95% (222 of 234). There was a higher sensitivity 97% (60 of 62) for detecting osteoporotic (dual-energy x-ray absorptiometry T-score \u2264 -2.5) lumbar vertebral fractures, implying easier detection, than for nonosteoporotic vertebral fractures (83% [39 of 47]). The artificial intelligence model also demonstrated better detection of lumbar vertebral fractures compared with detection of thoracic vertebral fractures based on the external dataset using various radiographic techniques. Based on the dataset for external validation, the overall accuracy, sensitivity, and specificity on bootstrapping method were 89%, 83%, and 95%, respectively.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The artificial intelligence model detected vertebral fractures on plain lateral radiographs with high accuracy, sensitivity, and specificity, especially for osteoporotic lumbar vertebral fractures (Genant Grades 2 and 3). The rapid reporting of results using this artificial intelligence model may improve the efficiency of diagnosing vertebral fractures. The testing model is available at http:\/\/140.113.114.104\/vght_demo\/corr\/. One or multiple plain lateral radiographs of the spine in the Digital Imaging and Communications in Medicine format can be uploaded to see the performance of the artificial intelligence model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Level of Evidence<\/jats:title>\n                    <jats:p>Level II, diagnostic study.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1097\/corr.0000000000001685","type":"journal-article","created":{"date-parts":[[2021,3,2]],"date-time":"2021-03-02T13:00:33Z","timestamp":1614690033000},"page":"1598-1612","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":95,"title":["Can a Deep-learning Model for the Automated Detection of Vertebral Fractures Approach the Performance Level of Human Subspecialists?"],"prefix":"10.1097","volume":"479","author":[{"given":"Yi-Chu","family":"Li","sequence":"first","affiliation":[{"name":"Institute of Data Science and Engineering, National Chiao Tung University, Hsinchu, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hung-Hsun","family":"Chen","sequence":"additional","affiliation":[{"name":"Center of Teaching and Learning Development, National Chiao Tung University, Hsinchu, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Henry","family":"Horng-Shing Lu","sequence":"additional","affiliation":[{"name":"Institute of Statistics, National Chiao Tung University, Hsinchu, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hung-Ta","family":"Hondar Wu","sequence":"additional","affiliation":[{"name":"School of Medicine, National Yang-Ming University, Taipei, Taiwan"},{"name":"Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming-Chau","family":"Chang","sequence":"additional","affiliation":[{"name":"School of Medicine, National Yang-Ming University, Taipei, Taiwan"},{"name":"Department of Orthopedics and Traumatology, Taipei Veterans General Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Po-Hsin","family":"Chou","sequence":"additional","affiliation":[{"name":"School of Medicine, National Yang-Ming University, Taipei, Taiwan"},{"name":"Department of Orthopedics and Traumatology, Taipei Veterans General Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2021,2,26]]},"reference":[{"key":"R1-20250505","doi-asserted-by":"crossref","first-page":"1410","DOI":"10.1097\/01.ta.0000197279.97113.0e","article-title":"Are plain radiographs of the spine necessary during evaluation after blunt trauma? Accuracy of screening torso computed tomography in thoracic\/lumbar spine fracture diagnosis","volume":"59","author":"Berry","year":"2005","journal-title":"J Trauma"},{"key":"R2-20250505","doi-asserted-by":"crossref","first-page":"1513","DOI":"10.1007\/s00198-005-1891-7","article-title":"Lateral vertebral assessment: a valuable technique to detect clinically significant vertebral fractures","volume":"16","author":"Binkley","year":"2005","journal-title":"Osteoporosis Int"},{"key":"R3-20250505","doi-asserted-by":"crossref","first-page":"788","DOI":"10.1148\/radiol.2017162100","article-title":"Vertebral body compression fractures and bone density: automated detection and classification on CT images","volume":"284","author":"Burns","year":"2017","journal-title":"Radiology"},{"key":"R4-20250505","doi-asserted-by":"crossref","first-page":"5469","DOI":"10.1007\/s00330-019-06167-y","article-title":"Application of a deep learning algorithm for detection and visualization of hip fractures on plain pelvic radiographs","volume":"29","author":"Cheng","year":"2019","journal-title":"Eur Radiol"},{"key":"R5-20250505","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":"Chung","year":"2018","journal-title":"Acta Orthop"},{"key":"R6-20250505","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1359\/JBMR.041214","article-title":"Underdiagnosis of vertebral fractures is a worldwide problem: the IMPACT study","volume":"20","author":"Delmas","year":"2005","journal-title":"J Bone Miner Res"},{"key":"R7-20250505","doi-asserted-by":"crossref","first-page":"1216","DOI":"10.1359\/JBMR.050314","article-title":"What proportion of incident radiographic vertebral deformities is clinically diagnosed and vice versa?","volume":"20","author":"Fink","year":"2005","journal-title":"J Bone Miner Res"},{"key":"R8-20250505","doi-asserted-by":"crossref","first-page":"895","DOI":"10.1007\/s00198-007-0530-x","article-title":"Back pain in osteoporotic vertebral fractures","volume":"19","author":"Francis","year":"2008","journal-title":"Osteoporosis Int"},{"key":"R9-20250505","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1007\/s003300000556","article-title":"Height of lumbar discs measured from radiographs compared with degeneration and height classified from MR images","volume":"11","author":"Frobin","year":"2001","journal-title":"Eur Radiol"},{"key":"R10-20250505","doi-asserted-by":"crossref","first-page":"394","DOI":"10.1080\/17453674.2019.1600125","article-title":"Artificial intelligence detection of distal radius fractures: a comparison between the convolutional neural network and professional assessments","volume":"90","author":"Gan","year":"2019","journal-title":"Acta Orthop"},{"key":"R11-20250505","doi-asserted-by":"crossref","first-page":"1137","DOI":"10.1002\/jbmr.5650080915","article-title":"Vertebral fracture assessment using a semiquantitative technique","volume":"8","author":"Genant","year":"1993","journal-title":"J Bone Miner Res"},{"key":"R12-20250505","doi-asserted-by":"crossref","first-page":"968","DOI":"10.1093\/rheumatology\/ket007","article-title":"Friend or foe: high bone mineral density on routine bone density scanning, a review of causes and management","volume":"52","author":"Gregson","year":"2013","journal-title":"Rheumatology (Oxford)"},{"key":"R13-20250505","doi-asserted-by":"crossref","first-page":"519","DOI":"10.1016\/j.rcl.2010.02.012","article-title":"Vertebral fracture","volume":"48","author":"Griffith","year":"2010","journal-title":"Radiol Clin North Am"},{"key":"R14-20250505","doi-asserted-by":"crossref","first-page":"142","DOI":"10.1016\/j.ejrad.2007.12.001","article-title":"Assessment of osteoporotic vertebral fractures using specialized workflow software for 6-point morphometry","volume":"70","author":"Guglielmi","year":"2009","journal-title":"Eur Radiol"},{"key":"R15-20250505","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":"Osteoporosis Int"},{"key":"R16-20250505","doi-asserted-by":"crossref","first-page":"2115","DOI":"10.1007\/s00586-012-2325-9","article-title":"Schmorl's nodes","volume":"21","author":"Kyere","year":"2012","journal-title":"Eur Spine J"},{"key":"R17-20250505","doi-asserted-by":"crossref","first-page":"949","DOI":"10.2214\/ajr.183.4.1830949","article-title":"Diagnosis of osteoporotic vertebral fractures: importance of recognition and description by radiologists","volume":"183","author":"Lenchik","year":"2004","journal-title":"AJR Am J Roentgenol"},{"key":"R18-20250505","first-page":"27","article-title":"Recognizing and reporting vertebral fractures: reducing the risk of future osteoporotic fractures","volume":"58","author":"Lentle","year":"2007","journal-title":"Can Assoc Radiol J"},{"key":"R19-20250505","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":"Olczak","year":"2017","journal-title":"Acta Orthop"},{"key":"R20-20250505","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"R21-20250505","doi-asserted-by":"crossref","first-page":"549","DOI":"10.1016\/S0140-6736(14)61347-7","article-title":"The burden of disease in older people and implications for health policy and practice","volume":"385","author":"Prince","year":"2015","journal-title":"Lancet"},{"key":"R22-20250505","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1007\/s00198-010-1195-4","article-title":"Severity of vertebral fracture and risk of hip fracture: a nested case-control study","volume":"22","author":"Puisto","year":"2011","journal-title":"Osteoporosis Int"},{"key":"R24-20250505","doi-asserted-by":"crossref","first-page":"430","DOI":"10.1097\/BRS.0b013e3181cdde47","article-title":"Spine computed tomography doses and cancer induction","volume":"35","author":"Richards","year":"2010","journal-title":"Spine"},{"key":"R25-20250505","doi-asserted-by":"crossref","first-page":"772","DOI":"10.21037\/qims.2016.12.04","article-title":"Classifying thoracolumbar fractures: role of quantitative imaging","volume":"6","author":"Ruiz Santiago","year":"2016","journal-title":"Quant Imaging Med Surg"},{"key":"R26-20250505","doi-asserted-by":"crossref","first-page":"e462","DOI":"10.5435\/JAAOS-D-17-00748","article-title":"Scheuermann's kyphosis: diagnosis, management, and selecting fusion levels","volume":"27","author":"Sardar","year":"2019","journal-title":"J Am Acad Orthop Surg"},{"key":"R28-20250505","doi-asserted-by":"crossref","first-page":"409","DOI":"10.1097\/BSD.0b013e3181568656","article-title":"The radiation exposure associated with cervical and lumbar spine radiographs","volume":"21","author":"Simpson","year":"2008","journal-title":"J Spinal Disord Tech"},{"key":"R29-20250505","doi-asserted-by":"crossref","first-page":"385","DOI":"10.3174\/ajnr.A5490","article-title":"Systematic radiation dose reduction in cervical spine CT of human cadaveric specimens: how low can we go?","volume":"39","author":"Tozakidou","year":"2018","journal-title":"Am J Neuroradiol"},{"key":"R30-20250505","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1016\/j.jemermed.2018.10.032","article-title":"Blunt thoracolumbar-spine trauma evaluation in the emergency department: a meta-analysis of diagnostic accuracy for history, physical examination, and imaging","volume":"56","author":"VandenBerg","year":"2019","journal-title":"J Emerg Med"},{"key":"R31-20250505","doi-asserted-by":"crossref","first-page":"871","DOI":"10.1007\/s00198-003-1461-9","article-title":"Clinical utility of dual-energy vertebral assessment (DVA)","volume":"14","author":"Vokes","year":"2003","journal-title":"Osteoporosis Int"},{"key":"R32-20250505","first-page":"205","article-title":"Vertebral compression fractures: a review of current management and multimodal therapy","volume":"6","author":"Wong","year":"2013","journal-title":"J Multidiscip Healthc"}],"container-title":["Clinical Orthopaedics &amp; Related Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.lww.com\/10.1097\/CORR.0000000000001685","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,5]],"date-time":"2025-05-05T13:17:53Z","timestamp":1746451073000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.lww.com\/10.1097\/CORR.0000000000001685"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,2,26]]},"references-count":30,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2021]]}},"URL":"https:\/\/doi.org\/10.1097\/corr.0000000000001685","relation":{},"ISSN":["0009-921X","1528-1132"],"issn-type":[{"value":"0009-921X","type":"print"},{"value":"1528-1132","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,2,26]]},"assertion":[{"value":"2020-09-18","name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-01-27","name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}