{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T15:56:35Z","timestamp":1786982195247,"version":"build-2736575974"},"reference-count":22,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2023,11,16]],"date-time":"2023-11-16T00:00:00Z","timestamp":1700092800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,11,16]],"date-time":"2023-11-16T00:00:00Z","timestamp":1700092800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Emma Louis Kessler Stiftung"},{"DOI":"10.13039\/501100006447","name":"Universit\u00e4t Z\u00fcrich","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100006447","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006447","name":"University of Zurich","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100006447","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Forensic Sci Med Pathol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Human or time resources can sometimes fall short in medical image diagnostics, and analyzing images in full detail can be a challenging task. With recent advances in artificial intelligence, an increasing number of systems have been developed to assist clinicians in their work. In this study, the objective was to train a model that can distinguish between various fracture types on different levels of hierarchical taxonomy and detect them on 2D-image representations of volumetric postmortem computed tomography (PMCT) data. We used a deep learning model based on the ResNet50 architecture that was pretrained on ImageNet data, and we used transfer learning to fine-tune it to our specific task. We trained our model to distinguish between \u201cdisplaced,\u201d \u201cnondisplaced,\u201d \u201cad latus,\u201d \u201cad longitudinem cum contractione,\u201d and \u201cad longitudinem cum distractione\u201d fractures. Radiographs with no fractures were correctly predicted in 95\u201399% of cases. Nondisplaced fractures were correctly predicted in 80\u201386% of cases. Displaced fractures of the \u201cad latus\u201d type were correctly predicted in 17\u201318% of cases. The other two displaced types of fractures, \u201cad longitudinem cum contractione\u201d and \u201cad longitudinem cum distractione,\u201d were correctly predicted in 70\u201375% and 64\u201375% of cases, respectively. The model achieved the best performance when the level of hierarchical taxonomy was high, while it had more difficulties when the level of hierarchical taxonomy was lower. Overall, deep learning techniques constitute a reliable solution for forensic pathologists and medical practitioners seeking to reduce workload.<\/jats:p>","DOI":"10.1007\/s12024-023-00751-x","type":"journal-article","created":{"date-parts":[[2023,11,15]],"date-time":"2023-11-15T23:34:53Z","timestamp":1700091293000},"page":"1208-1214","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Classification of rib fracture types from postmortem computed tomography images using deep learning"],"prefix":"10.1007","volume":"20","author":[{"given":"Victor","family":"Ibanez","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dario","family":"Jucker","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lars C.","family":"Ebert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sabine","family":"Franckenberg","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6492-9298","authenticated-orcid":false,"given":"Akos","family":"Dobay","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,11,16]]},"reference":[{"key":"751_CR1","doi-asserted-by":"publisher","first-page":"975","DOI":"10.1097\/00005373-199412000-00018","volume":"37","author":"DW Ziegler","year":"1994","unstructured":"Ziegler DW, Agarwal NN. The morbidity and mortality of rib fractures. J Trauma. 1994;37:975\u20139.","journal-title":"J Trauma"},{"key":"751_CR2","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1016\/S1010-7940(03)00256-2","volume":"24","author":"M Sirmali","year":"2003","unstructured":"Sirmali M, Turut H, Topcu S, et al. A comprehensive analysis of traumatic rib fractures: Morbidity, mortality and management. Eur J Cardiothorac Surg. 2003;24:133\u20138.","journal-title":"Eur J Cardiothorac Surg"},{"key":"751_CR3","first-page":"235","volume":"44","author":"J Crandall","year":"2000","unstructured":"Crandall J, Kent R, Patrie J, et al. Rib fracture patterns and radiologic detection\u2013a restraint-based comparison. Annu Proc Assoc Adv Automot Med. 2000;44:235\u201359.","journal-title":"Annu Proc Assoc Adv Automot Med"},{"key":"751_CR4","doi-asserted-by":"publisher","DOI":"10.1136\/tsaco-2020-000560","volume":"5","author":"V Pedersen","year":"2020","unstructured":"Pedersen V, Lampart A, Bingisser R, et al. Accuracy of plain radiography in detecting fractures in older individuals after low-energy falls: Current evidence. Trauma Surg Acute Care Open. 2020;5:e000560.","journal-title":"Trauma Surg Acute Care Open"},{"key":"751_CR5","first-page":"351","volume":"16","author":"M Chardoli","year":"2013","unstructured":"Chardoli M, Hasan-Ghaliaee T, Akbari H, et al. Accuracy of chest radiography versus chest computed tomography in hemodynamically stable patients with blunt chest trauma. Chin J Traumatol. 2013;16:351\u20134.","journal-title":"Chin J Traumatol"},{"key":"751_CR6","doi-asserted-by":"publisher","first-page":"2078","DOI":"10.1001\/archinternmed.2009.427","volume":"169","author":"R Smith-Bindman","year":"2009","unstructured":"Smith-Bindman R, Lipson J, Marcus R, et al. Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Arch Intern Med. 2009;169:2078\u201386.","journal-title":"Arch Intern Med"},{"key":"751_CR7","doi-asserted-by":"crossref","unstructured":"Bolliger SA, Thali MJ. Imaging and virtual autopsy: Looking back and forward. Philos Trans R Soc Lond B Biol Sci\u00a02015;370","DOI":"10.1098\/rstb.2014.0253"},{"key":"751_CR8","doi-asserted-by":"publisher","first-page":"11591","DOI":"10.1073\/pnas.1806905115","volume":"115","author":"R Lindsey","year":"2018","unstructured":"Lindsey R, Daluiski A, Chopra S, et al. Deep neural network improves fracture detection by clinicians. Proc Natl Acad Sci U S A. 2018;115:11591\u20136.","journal-title":"Proc Natl Acad Sci U S A"},{"key":"751_CR9","doi-asserted-by":"publisher","first-page":"108925","DOI":"10.1016\/j.ejrad.2020.108925","volume":"126","author":"C Bluthgen","year":"2020","unstructured":"Bluthgen C, Becker AS, Vittoria de Martini I, et al. Detection and localization of distal radius fractures: Deep learning system versus radiologists. Eur J Radiol. 2020;126:108925.","journal-title":"Eur J Radiol"},{"key":"751_CR10","doi-asserted-by":"publisher","first-page":"672","DOI":"10.1007\/s10278-018-0167-7","volume":"32","author":"G Kitamura","year":"2019","unstructured":"Kitamura G, Chung CY, Moore BE 2nd. Ankle fracture detection utilizing a convolutional neural network ensemble implemented with a small sample, de novo training, and multiview incorporation. J Digit Imaging. 2019;32:672\u20137.","journal-title":"J Digit Imaging"},{"key":"751_CR11","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1148\/radiol.2015142346","volume":"278","author":"JE Burns","year":"2016","unstructured":"Burns JE, Yao JH, Munoz H, et al. Automated detection, localization, and classification of traumatic vertebral body fractures in the thoracic and lumbar spine at CT. Radiology. 2016;278:64\u201373.","journal-title":"Radiology"},{"key":"751_CR12","doi-asserted-by":"publisher","first-page":"891","DOI":"10.3348\/kjr.2019.0653","volume":"21","author":"T Weikert","year":"2020","unstructured":"Weikert T, Noordtzij LA, Bremerich J, et al. Assessment of a deep learning algorithm for the detection of rib fractures on whole-body trauma computed tomography. Korean J Radiol. 2020;21:891\u20139.","journal-title":"Korean J Radiol"},{"key":"751_CR13","unstructured":"Choi J, Edamadaka S, Brown D. Deep learning to automate identification and characterization of rib fractures on chest computed tomography scans. Stanford CS class CS231n: Convolutional Neural Networks for Visual Recognition. Final report; 2022. p. 1\u20139. http:\/\/cs231n.stanford.edu\/reports\/2022\/pdfs\/68.pdf. Accessed 10 Nov 2023."},{"key":"751_CR14","first-page":"234","volume-title":"Medical image computing and computer-assisted intervention \u2013 MICCAI 2015","author":"O Ronneberger","year":"2015","unstructured":"Ronneberger O, Fischer P, Brox T, et al. U-Net: Convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells WM, et al., editors. Medical image computing and computer-assisted intervention \u2013 MICCAI 2015. Cham: Springer International Publishing; 2015. p. 234\u201341."},{"key":"751_CR15","doi-asserted-by":"crossref","unstructured":"Wang X, Wang Y. Composite attention residual U-Net for rib fracture detection. Entropy (Basel). 2023;25.","DOI":"10.3390\/e25030466"},{"key":"751_CR16","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1186\/s12880-023-00975-x","volume":"23","author":"J Wu","year":"2023","unstructured":"Wu J, Liu N, Li X, et al. Convolutional neural network for detecting rib fractures on chest radiographs: A feasibility study. BMC Med Imaging. 2023;23:18.","journal-title":"BMC Med Imaging"},{"key":"751_CR17","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1007\/s00414-012-0686-6","volume":"126","author":"C Egger","year":"2012","unstructured":"Egger C, Vaucher P, Doenz F, et al. Development and validation of a postmortem radiological alteration index: The RA-index. Int J Legal Med. 2012;126:559\u201366.","journal-title":"Int J Legal Med"},{"key":"751_CR18","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1007\/s12024-014-9555-6","volume":"10","author":"PM Flach","year":"2014","unstructured":"Flach PM, Gascho D, Schweitzer W, et al. Imaging in forensic radiology: An illustrated guide for postmortem computed tomography technique and protocols. Forensic Sci Med Pathol. 2014;10:583\u2013606.","journal-title":"Forensic Sci Med Pathol"},{"key":"751_CR19","doi-asserted-by":"publisher","first-page":"1865","DOI":"10.1007\/s00330-015-3598-2","volume":"25","author":"H Ringl","year":"2015","unstructured":"Ringl H, Lazar M, Topker M, et al. The ribs unfolded - a CT visualization algorithm for fast detection of rib fractures: Effect on sensitivity and specificity in trauma patients. Eur Radiol. 2015;25:1865\u201374.","journal-title":"Eur Radiol"},{"key":"751_CR20","first-page":"770","volume":"2016","author":"K He","year":"2016","unstructured":"He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. IEEE Conf Comput Vis Pattern Recognit (CVPR). 2016;2016:770\u20138.","journal-title":"IEEE Conf Comput Vis Pattern Recognit (CVPR)"},{"key":"751_CR21","first-page":"265","volume-title":"Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation","author":"M Abadi","year":"2016","unstructured":"Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. TensorFlow: A system for large-scale machine learning. In: Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation. USENIX Association; 2016. p. 265\u201383."},{"key":"751_CR22","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1007\/s12024-021-00431-8","volume":"18","author":"V Ibanez","year":"2022","unstructured":"Ibanez V, Gunz S, Erne S, et al. RiFNet: Automated rib fracture detection in postmortem computed tomography. Forensic Sci Med Pathol. 2022;18:20\u20139.","journal-title":"Forensic Sci Med Pathol"}],"container-title":["Forensic Science, Medicine and Pathology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s12024-023-00751-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s12024-023-00751-x\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s12024-023-00751-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T07:13:53Z","timestamp":1738566833000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s12024-023-00751-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,16]]},"references-count":22,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,12]]}},"alternative-id":["751"],"URL":"https:\/\/doi.org\/10.1007\/s12024-023-00751-x","relation":{},"ISSN":["1556-2891"],"issn-type":[{"value":"1556-2891","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,16]]},"assertion":[{"value":"5 November 2023","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 November 2023","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This retrospective study does not fall within the scope of the Human Research Act (HRA). The Cantonal Ethics Board of the Canton of Zurich has waived the need for an authorization (waiver document KEK ZH-No. 15\u20130686).","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}]}}