{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T19:44:17Z","timestamp":1791488657897,"version":"4.3.4"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2020,4,17]],"date-time":"2020-04-17T00:00:00Z","timestamp":1587081600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,4,17]],"date-time":"2020-04-17T00:00:00Z","timestamp":1587081600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100011699","name":"Siemens Healthineers","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100011699","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol Exp"],"published-print":{"date-parts":[[2020,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n<jats:title>Background<\/jats:title>\n<jats:p>Automatically detecting and quantifying pneumothorax on chest computed tomography (CT) may impact clinical decision-making. Machine learning methods published so far struggle with the heterogeneity of technical parameters and the presence of additional pathologies, highlighting the importance of stable algorithms.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Methods<\/jats:title>\n<jats:p>A deep residual UNet was developed and evaluated\u00a0for automated, volume-level pneumothorax grading (<jats:italic>i.e.,<\/jats:italic> labelling a volume whether a pneumothorax was present or not), and pixel-level classification (<jats:italic>i.e.,<\/jats:italic> segmentation and quantification of pneumothorax), on a retrospective series of routine chest CT data. Ground truth annotations were provided by radiologists. The fully automated pixel-level pneumothorax segmentation method was trained using 43 chest CT scans and evaluated on 9 chest CT scans with pixel-level annotation basis and 567 chest CT scans on a volume-level basis.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Results<\/jats:title>\n<jats:p>This method achieved a receiver operating characteristic area under the curve (AUC) of 0.98, an average precision of 0.97, and a Dice similarity coefficient (DSC) of 0.94. This segmentation performance resulted to be similar to the inter-rater segmentation accuracy of two radiologists, who achieved a DSC of 0.92. The comparison of manual and automated pneumothorax quantification yielded a Pearson correlation coefficient of 0.996. The volume-level pneumothorax grading accuracy was evaluated on 567 chest CT scans and yielded an AUC of 0.98 and an average precision of 0.95.<\/jats:p>\n<\/jats:sec><jats:sec>\n<jats:title>Conclusions<\/jats:title>\n<jats:p>We proposed a deep learning method for the detection and quantification of pneumothorax in heterogeneous routine clinical data that may facilitate the automated triage of urgent examinations and enable treatment decision support.<\/jats:p>\n<\/jats:sec>","DOI":"10.1186\/s41747-020-00152-7","type":"journal-article","created":{"date-parts":[[2020,4,17]],"date-time":"2020-04-17T12:02:34Z","timestamp":1587124954000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":28,"title":["Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography"],"prefix":"10.1186","volume":"4","author":[{"given":"Sebastian","family":"R\u00f6hrich","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas","family":"Schlegl","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Constanze","family":"Bardach","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Helmut","family":"Prosch","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Georg","family":"Langs","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2020,4,17]]},"reference":[{"key":"152_CR1","doi-asserted-by":"publisher","first-page":"12215","DOI":"10.1038\/srep12215","volume":"5","author":"RHHM Philipsen","year":"2015","unstructured":"Philipsen RHHM, S\u00e1nchez CI, Maduskar P et al (2015) Automated chest-radiography as a triage for Xpert testing in resource-constrained settings: a prospective study of diagnostic accuracy and costs. Sci Rep 5:12215. https:\/\/doi.org\/10.1038\/srep12215","journal-title":"Sci Rep"},{"key":"152_CR2","doi-asserted-by":"publisher","unstructured":"Grunwald IQ, Kulikovski J, Reith W et al (2019) Collateral automation for triage in stroke: evaluating automated scoring of collaterals in acute stroke on computed tomography scans. Cerebrovasc Dis 1\u20136. https:\/\/doi.org\/10.1159\/000500076","DOI":"10.1159\/000500076"},{"key":"152_CR3","doi-asserted-by":"publisher","first-page":"769","DOI":"10.1055\/s-0030-1269837","volume":"31","author":"D Haynes","year":"2010","unstructured":"Haynes D, Baumann MH (2010) Management of pneumothorax. Semin Respir Crit Care Med 31:769\u2013780. https:\/\/doi.org\/10.1055\/s-0030-1269837","journal-title":"Semin Respir Crit Care Med"},{"key":"152_CR4","doi-asserted-by":"publisher","unstructured":"MacDuff A, Arnold A, Harvey J, BTS Pleural Disease Guideline Group (2010) Management of spontaneous pneumothorax: British Thoracic Society Pleural Disease Guideline 2010. Thorax 65 Suppl 2:ii18\u201331. https:\/\/doi.org\/10.1136\/thx.2010.136986","DOI":"10.1136\/thx.2010.136986"},{"key":"152_CR5","doi-asserted-by":"publisher","unstructured":"Hefny AF, Kunhivalappil FT, Matev N, Avila NA, Bashir MO, Abu-Zidan FM (2018) Management of computed tomography-detected pneumothorax in patients with blunt trauma: experience from a community-based hospital. Singapore Med J 59:150\u2013154. https:\/\/doi.org\/10.11622\/smedj.2017074","DOI":"10.11622\/smedj.2017074"},{"key":"152_CR6","doi-asserted-by":"publisher","unstructured":"CCai W, Lee JG, Fikry K, Yoshida H, Novelline R, de Moya M (2012) MDCT quantification is the dominant parameter in decision-making regarding chest tube drainage for stable patients with traumatic pneumothorax. Comput Med Imaging Graph 36:375\u2013386. https:\/\/doi.org\/10.1016\/j.compmedimag.2012.03.005","DOI":"10.1016\/j.compmedimag.2012.03.005"},{"key":"152_CR7","doi-asserted-by":"publisher","first-page":"590","DOI":"10.1378\/chest.119.2.590","volume":"119","author":"MH Baumann","year":"2001","unstructured":"Baumann MH, Strange C, Heffner JE et al (2001) Management of spontaneous pneumothorax: an American College of Chest Physicians Delphi consensus statement. Chest 119:590\u2013602","journal-title":"Chest"},{"key":"152_CR8","doi-asserted-by":"publisher","first-page":"946","DOI":"10.1016\/j.chest.2017.10.015","volume":"153","author":"SP Walker","year":"2018","unstructured":"Walker SP, Barratt SL, Thompson J, Maskell NA (2018) Conservative management in traumatic pneumothoraces: an observational study. Chest 153:946\u2013953. https:\/\/doi.org\/10.1016\/j.chest.2017.10.015","journal-title":"Chest"},{"key":"152_CR9","doi-asserted-by":"publisher","unstructured":"Do S, Salvaggio K, Gupta S, Kalra M, Ali NU, Pien H (2012) Automated quantification of pneumothorax in CT. Comput Math Methods Med 2012:1\u20137. https:\/\/doi.org\/10.1155\/2012\/736320","DOI":"10.1155\/2012\/736320"},{"key":"152_CR10","doi-asserted-by":"publisher","first-page":"830","DOI":"10.2214\/AJR.08.1339","volume":"192","author":"W Cai","year":"2009","unstructured":"Cai W, Tabbara M, Takata N et al (2009) MDCT for automated detection and measurement of pneumothoraces in trauma patients. AJR Am J Roentgenol 192:830\u2013836. https:\/\/doi.org\/10.2214\/AJR.08.1339","journal-title":"AJR Am J Roentgenol"},{"key":"152_CR11","doi-asserted-by":"publisher","unstructured":"Cai W, Lee EY, Vij A, Mahmood SA, Yoshida H (2011) MDCT for computerized volumetry of pneumothoraces in pediatric patients. Acad Radiol 18:315\u2013323. https:\/\/doi.org\/10.1016\/j.acra.2010.11.008","DOI":"10.1016\/j.acra.2010.11.008"},{"key":"152_CR12","doi-asserted-by":"publisher","first-page":"2293","DOI":"10.1109\/tmi.2014.2337057","volume":"33","author":"A Mansoor","year":"2014","unstructured":"Mansoor A, Bagci U, Xu Z et al (2014) A generic approach to pathological lung segmentation. IEEE Trans Med Imaging 33:2293\u20132310. https:\/\/doi.org\/10.1109\/tmi.2014.2337057","journal-title":"IEEE Trans Med Imaging"},{"key":"152_CR13","doi-asserted-by":"publisher","first-page":"108692","DOI":"10.1016\/j.ejrad.2019.108692","volume":"120","author":"X Li","year":"2019","unstructured":"Li X, Thrall JH, Digumarthy SR et al (2019) Deep learning-enabled system for rapid pneumothorax screening on chest CT. Eur J Radiol 120:108692. https:\/\/doi.org\/10.1016\/j.ejrad.2019.108692","journal-title":"Eur J Radiol"},{"key":"152_CR14","doi-asserted-by":"publisher","first-page":"e1002697","DOI":"10.1371\/journal.pmed.1002697","volume":"15","author":"AG Taylor","year":"2018","unstructured":"Taylor AG, Mielke C, Mongan J (2018) Automated detection of moderate and large pneumothorax on frontal chest X-rays using deep convolutional neural networks: A retrospective study. PLoS Med 15:e1002697. https:\/\/doi.org\/10.1371\/journal.pmed.1002697","journal-title":"PLoS Med"},{"key":"152_CR15","doi-asserted-by":"publisher","unstructured":"Chan YH, Zeng YZ, Wu HC, Wu MC, Sun HM (2018) Effective pneumothorax detection for chest x-ray images using local binary pattern and support vector machine. J Healthc Eng 2018:2908517. https:\/\/doi.org\/10.1155\/2018\/2908517","DOI":"10.1155\/2018\/2908517"},{"key":"152_CR16","doi-asserted-by":"publisher","first-page":"1116","DOI":"10.1016\/j.neuroimage.2006.01.015","volume":"31","author":"PA Yushkevich","year":"2006","unstructured":"Yushkevich PA, Piven J, Hazlett HC et al (2006) User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31:1116\u20131128. https:\/\/doi.org\/10.1016\/j.neuroimage.2006.01.015","journal-title":"Neuroimage"},{"key":"152_CR17","doi-asserted-by":"crossref","unstructured":"Noh H, Hong S, Han B (2015) Learning deconvolution network for semantic segmentation. 2015 IEEE International Conference on Computer Vision (ICCV). https:\/\/www.cv-foundation.org\/openaccess\/content_iccv_2015\/papers\/Noh_Learning_Deconvolution_Network_ICCV_2015_paper.pdf","DOI":"10.1109\/ICCV.2015.178"},{"key":"152_CR18","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional networks for biomedical image segmentation. Lecture Notes Comp Sci:234\u2013241 https:\/\/arxiv.org\/abs\/1505.04597","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"152_CR19","unstructured":"Abadi M, Agarwal A, Barham P et al (2016) TensorFlow: large-scale machine learning on heterogeneous distributed systems. https:\/\/arxiv.org\/pdf\/1603.04467.pdf"},{"key":"152_CR20","unstructured":"Pawlowski N, Ktena SI, Lee MCH et al (2017) DLTK: State of the art reference implementations for deep learning on medical images. https:\/\/arxiv.org\/abs\/1711.06853"},{"key":"152_CR21","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","volume":"3","author":"WJ Youden","year":"1950","unstructured":"Youden WJ (1950) Index for rating diagnostic tests. Cancer 3:32\u201335. https:\/\/doi.org\/10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","journal-title":"Cancer"},{"key":"152_CR22","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1378\/chest.103.1.26","volume":"103","author":"O Engdahl","year":"1993","unstructured":"Engdahl O, Toft T, Boe J (1993) Chest radiograph\u2014a poor method for determining the size of a pneumothorax. Chest 103:26\u201329. https:\/\/doi.org\/10.1378\/chest.103.1.26","journal-title":"Chest"},{"key":"152_CR23","doi-asserted-by":"publisher","first-page":"1830","DOI":"10.1016\/j.rmed.2008.07.026","volume":"102","author":"AM Kelly","year":"2008","unstructured":"Kelly AM, Druda D (2008) Comparison of size classification of primary spontaneous pneumothorax by three international guidelines: a case for international consensus? Respir Med 102:1830\u20131832. https:\/\/doi.org\/10.1016\/j.rmed.2008.07.026","journal-title":"Respir Med"},{"key":"152_CR24","doi-asserted-by":"crossref","unstructured":"Henry M, Arnold T, Harvey J (2003) BTS guidelines for the management of spontaneous pneumothorax. Thorax 58 Suppl 2:ii39-52. http:\/\/dx.doi.org\/10.1136\/thx.58.suppl_2.ii39","DOI":"10.1136\/thx.58.suppl_2.ii39"},{"key":"152_CR25","doi-asserted-by":"publisher","unstructured":"O\u2019Rourke JP, Yee ES (1989) Civilian spontaneous pneumothorax. Treatment options and long-term results. Chest 96:1302\u20131306. https:\/\/doi.org\/10.1378\/chest.96.6.1302","DOI":"10.1378\/chest.96.6.1302"},{"key":"152_CR26","doi-asserted-by":"publisher","unstructured":"Kroft LJM, van der Velden L, Gir\u00f3n IH, Roelofs JJH, de Roos A, Geleijns J (2019) Added value of ultra\u2013low-dose computed tomography, dose Equivalent to chest x-ray radiography, for diagnosing chest pathology. J Thorac Imaging 34:179\u2013186. https:\/\/doi.org\/10.1097\/rti.0000000000000404","DOI":"10.1097\/rti.0000000000000404"}],"container-title":["European Radiology Experimental"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-020-00152-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s41747-020-00152-7\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-020-00152-7.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,4,16]],"date-time":"2021-04-16T23:14:57Z","timestamp":1618614897000},"score":1,"resource":{"primary":{"URL":"https:\/\/eurradiolexp.springeropen.com\/articles\/10.1186\/s41747-020-00152-7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,4,17]]},"references-count":26,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,12]]}},"alternative-id":["152"],"URL":"https:\/\/doi.org\/10.1186\/s41747-020-00152-7","relation":{},"ISSN":["2509-9280"],"issn-type":[{"value":"2509-9280","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,4,17]]},"assertion":[{"value":"13 September 2019","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 March 2020","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 April 2020","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The local ethics committee of the Medical University of Vienna approved the retrospective analysis of the imaging data for the study (approval number 1154\/2014).","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"SR is doing consulting activities for contextflow GmbH. HP receives speaker fees from Boehringer-Ingelheim, Roche, Novartis, MSD, BMS, GSK, Chiesi, and AstraZeneca, and is a research support for Boehringer-Ingelheim. GL is a shareholder\/co-founder of contextflow GmbH, receives speaker fees from Roche, Boehringer-Ingelheim, and Siemens, and receives research support from Siemens, Novartis, IBM, and NVIDIA. TS and CB declare that they have no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"26"}}