{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T08:16:38Z","timestamp":1791965798046,"version":"4.3.4"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,1,25]],"date-time":"2023-01-25T00:00:00Z","timestamp":1674604800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,1,25]],"date-time":"2023-01-25T00:00:00Z","timestamp":1674604800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Crit Care"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Chest radiographs are routinely performed in intensive care unit (ICU) to confirm the correct position of an endotracheal tube (ETT) relative to the carina. However, their interpretation is often challenging and requires substantial time and expertise. The aim of this study was to propose an externally validated deep learning model with uncertainty quantification and image segmentation for the automated assessment of ETT placement on ICU chest radiographs.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>The CarinaNet model was constructed by applying transfer learning to the RetinaNet model using an internal dataset of ICU chest radiographs. The accuracy of the model in predicting the position of the ETT tip and carina was externally validated using a dataset of 200 images extracted from the MIMIC-CXR database. Uncertainty quantification was performed using the level of confidence in the ETT\u2013carina distance prediction. Segmentation of the ETT was carried out using edge detection and pixel clustering.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>The interrater agreement was 0.18\u00a0cm for the ETT tip position, 0.58\u00a0cm for the carina position, and 0.60\u00a0cm for the ETT\u2013carina distance. The mean absolute error of the model on the external test set was 0.51\u00a0cm for the ETT tip position prediction, 0.61\u00a0cm for the carina position prediction, and 0.89\u00a0cm for the ETT\u2013carina distance prediction. The assessment of ETT placement was improved by complementing the human interpretation of chest radiographs with the CarinaNet model.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusions<\/jats:title>\n                <jats:p>The CarinaNet model is an efficient and generalizable deep learning algorithm for the automated assessment of ETT placement on ICU chest radiographs. Uncertainty quantification can bring the attention of intensivists to chest radiographs that require an experienced human interpretation. Image segmentation provides intensivists with chest radiographs that are quickly interpretable and allows them to immediately assess the validity of model predictions. The CarinaNet model is ready to be evaluated in clinical studies.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Graphical Abstract<\/jats:title>\n                \n              <\/jats:sec>","DOI":"10.1186\/s13054-023-04320-0","type":"journal-article","created":{"date-parts":[[2023,1,25]],"date-time":"2023-01-25T15:04:08Z","timestamp":1674659048000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":16,"title":["Image augmentation and automated measurement of endotracheal-tube-to-carina distance on chest radiographs in intensive care unit using a deep learning model with external validation"],"prefix":"10.1186","volume":"27","author":[{"given":"Matthieu","family":"Oliver","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Am\u00e9lie","family":"Renou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nicolas","family":"Allou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lucas","family":"Moscatelli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cyril","family":"Ferdynus","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jer\u00f4me","family":"Allyn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,1,25]]},"reference":[{"key":"4320_CR1","doi-asserted-by":"publisher","first-page":"1687","DOI":"10.1016\/S0140-6736(09)61459-8","volume":"374","author":"G Hejblum","year":"2009","unstructured":"Hejblum G, Chalumeau-Lemoine L, Ioos V, et al. Comparison of routine and on-demand prescription of chest radiographs in mechanically ventilated adults: a multicentre, cluster-randomised, two-period crossover study. Lancet. 2009;374:1687\u201393.","journal-title":"Lancet"},{"key":"4320_CR2","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/S0750-7658(03)00127-8","volume":"22","author":"C Colin","year":"2003","unstructured":"Colin C. Prise en charge des voies a\u00e9riennes en anesth\u00e9sie adulte \u00e0 l\u2019exception de l\u2019intubation difficile. Annales Fran\u00e7aises d\u2019Anesth\u00e9sie et de R\u00e9animation. 2003;22:3\u201317.","journal-title":"Annales Fran\u00e7aises d\u2019Anesth\u00e9sie et de R\u00e9animation."},{"key":"4320_CR3","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1016\/j.bja.2017.10.021","volume":"120","author":"A Higgs","year":"2017","unstructured":"Higgs A, Mcgrath B, Goddard C, et al. Guidelines for the management of tracheal intubation in critically ill adults. Br J Anaesth. 2017;120:323\u201352.","journal-title":"Br J Anaesth"},{"key":"4320_CR4","doi-asserted-by":"publisher","first-page":"R68","DOI":"10.1186\/cc11321","volume":"16","author":"A Ganapathy","year":"2012","unstructured":"Ganapathy A, Adhikari N, Spiegelman J, Scales D. Routine chest x-rays in intensive care units: a systemic review and meta-analysis. Crit Care. 2012;16:R68.","journal-title":"Crit Care"},{"key":"4320_CR5","doi-asserted-by":"publisher","first-page":"1342","DOI":"10.1038\/s41591-018-0107-6","volume":"24","author":"J Fauw","year":"2018","unstructured":"Fauw J, Ledsam J, Romera-Paredes B, et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat Med. 2018;24:1342\u201350.","journal-title":"Nat Med"},{"key":"4320_CR6","doi-asserted-by":"publisher","first-page":"2402","DOI":"10.1001\/jama.2016.17216","volume":"316","author":"V Gulshan","year":"2016","unstructured":"Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016;316:2402\u201310.","journal-title":"JAMA"},{"key":"4320_CR7","doi-asserted-by":"publisher","first-page":"449","DOI":"10.1007\/s10278-017-9983-4","volume":"30","author":"Z Akkus","year":"2017","unstructured":"Akkus Z, Galimzianova A, Hoogi A, Rubin D, Erickson B. Deep learning for brain MRI segmentation: state of the art and future directions. J Digit Imaging. 2017;30:449\u201359.","journal-title":"J Digit Imaging"},{"key":"4320_CR8","doi-asserted-by":"publisher","first-page":"1616","DOI":"10.1007\/s00330-018-5674-x","volume":"29","author":"P Savadjiev","year":"2018","unstructured":"Savadjiev P, Chong J, Dohan A, et al. Demystification of AI-driven medical image interpretation: past, present and future. Eur Radiol. 2018;29:1616\u201324.","journal-title":"Eur Radiol"},{"key":"4320_CR9","first-page":"1","volume":"31","author":"H Lee","year":"2017","unstructured":"Lee H, Mansouri M, Tajmir S, Lev M, Do S. A deep-learning system for fully-automated peripherally inserted central catheter (PICC) tip detection. J Digit Imaging. 2017;31:1\u201310.","journal-title":"J Digit Imaging"},{"key":"4320_CR10","unstructured":"Zhimin H, Jing Z, inventors; Carestream\u00a0Health Inc, assignee. Method for detecting anatomical structures. US 8,577,108 B2; 2013."},{"key":"4320_CR11","unstructured":"Laoa Z, Zhengb X, Zouc Q. Carina landmark detection in ICU images via integrating geometrical and thoracic anatomy based features. In: 21st international conference on pattern recognition. 2012; p. 5\u20138."},{"key":"4320_CR12","first-page":"784","volume-title":"Endotracheal tube detection and segmentation in chest radiographs using synthetic data","author":"M Frid-Adar","year":"2019","unstructured":"Frid-Adar M, Amer R, Greenspan H. Medical image computing and computer assisted intervention. In: Endotracheal tube detection and segmentation in chest radiographs using synthetic data. Cham: Springer; 2019. p. 784\u201392."},{"key":"4320_CR13","doi-asserted-by":"publisher","first-page":"898","DOI":"10.1007\/s10278-021-00463-0","volume":"34","author":"S Kara","year":"2021","unstructured":"Kara S, Akers J, Chang P. Identification and Localization of endotracheal tube on chest radiographs using a cascaded convolutional neural network approach. J Digit Imaging. 2021;34:898\u2013904.","journal-title":"J Digit Imaging"},{"key":"4320_CR14","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2020200026","volume":"3","author":"P Lakhani","year":"2021","unstructured":"Lakhani P, Flanders A, Gorniak R. Endotracheal tube position assessment on chest radiographs using deep learning. Radiol Artif Intell. 2021;3: e200026.","journal-title":"Radiol Artif Intell"},{"key":"4320_CR15","doi-asserted-by":"crossref","unstructured":"Johnson A, Pollard T, Berkowitz S, et\u00a0al. MIMIC-CXR: a large publicly available database of labeled chest radiographs. Scientific Data. 2019;6.","DOI":"10.1038\/s41597-019-0322-0"},{"key":"4320_CR16","doi-asserted-by":"publisher","first-page":"1364","DOI":"10.1038\/s41591-020-1034-x","volume":"26","author":"X Liu","year":"2020","unstructured":"Liu X, Cruz Rivera S, Moher D, et al. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: The CONSORT-AI Extension. Nat Med. 2020;26:1364\u201374.","journal-title":"Nat Med"},{"key":"4320_CR17","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1111\/j.1365-2044.2009.06007.x","volume":"64","author":"W Geisser","year":"2009","unstructured":"Geisser W, Maybauer D, Wolff H, Pfenninger E, Maybauer M. Radiological validation of tracheal tube insertion depth in out-hospital and in-hospital emergency patients. Anaesthesia. 2009;64:973\u20137.","journal-title":"Anaesthesia"},{"key":"4320_CR18","doi-asserted-by":"crossref","unstructured":"Lin TY, Goyal P, Girshick R, He K, Dollar P. Focal Loss for Dense Object Detection. IEEE International Conference on Computer Vision. 2017; p. 2999\u20133007.","DOI":"10.1109\/ICCV.2017.324"},{"key":"4320_CR19","first-page":"740","volume":"05","author":"TY Lin","year":"2014","unstructured":"Lin TY, Maire M, Belongie S, et al. Microsoft COCO: common objects in context. European conference on computer vision. 2014;05:740\u201355.","journal-title":"European conference on computer vision"},{"key":"4320_CR20","doi-asserted-by":"publisher","first-page":"531","DOI":"10.3390\/rs11050531","volume":"11","author":"Y Wang","year":"2019","unstructured":"Wang Y, Wang C, Zhang H, Dong Y, Wei S. Automatic ship detection based on RetinaNet using multi-resolution Gaofen-3 imagery. Remote Sens. 2019;11:531.","journal-title":"Remote Sens"},{"key":"4320_CR21","first-page":"68","volume":"32","author":"H Wei","year":"2020","unstructured":"Wei H, Zhao Y, Dong J. Cooling tower detection based on the improved RetinaNet. Remote Sens Nat Resour. 2020;32:68\u201373.","journal-title":"Remote Sens Nat Resour"},{"key":"4320_CR22","doi-asserted-by":"crossref","unstructured":"Zlocha M, Dou Q, Glocker B; Springer. Improving RetinaNet for CT lesion detection with dense masks from weak RECIST labels. In: International conference on medical image computing and computer-assisted intervention. 2019; p. 402\u201310.","DOI":"10.1007\/978-3-030-32226-7_45"},{"key":"4320_CR23","first-page":"1","volume":"2020","author":"M Yang","year":"2020","unstructured":"Yang M, Xiao X, Liu Z, et al. Deep RetinaNet for dynamic left ventricle detection in multiview echocardiography classification. Sci Program. 2020;2020:1\u20136.","journal-title":"Sci Program"},{"key":"4320_CR24","doi-asserted-by":"crossref","unstructured":"Mao L, Yumeng T, Lina C. Pneumonia detection in chest X-rays: a deep learning approach based on ensemble RetinaNet and Mask R-CNN. In: Eighth international conference on advanced Cloud and Big Data (CBD). 2020; p. 213\u201318.","DOI":"10.1109\/CBD51900.2020.00046"},{"key":"4320_CR25","doi-asserted-by":"crossref","unstructured":"Lin TY, Doll\u00e1r P, Girshick R, He K, Hariharan B, Belongie S. Feature Pyramid Networks for Object Detection. In: IEEE conference on computer vision and pattern recognition. 2017; p. 936\u201344.","DOI":"10.1109\/CVPR.2017.106"},{"issue":"4","key":"4320_CR26","doi-asserted-by":"publisher","first-page":"501","DOI":"10.1109\/TMI.2004.825627","volume":"23","author":"J Staal","year":"2004","unstructured":"Staal J, Abramoff MD, Niemeijer M, Viergever MA, van Ginneken B. Ridge-based vessel segmentation in color images of the retina. IEEE Trans Med Imaging. 2004;23(4):501\u20139.","journal-title":"IEEE Trans Med Imaging"},{"key":"4320_CR27","unstructured":"Ester M, Kriegel HP, Sander J, Xu X. A density-based algorithm for discovering clusters in large spatial databases with noise. In: Proceedings of the second international conference on knowledge discovery and data mining. 1996; p. 226\u201331."},{"issue":"1","key":"4320_CR28","doi-asserted-by":"publisher","first-page":"e15","DOI":"10.1016\/S2589-7500(19)30011-1","volume":"05","author":"K Gruber","year":"2019","unstructured":"Gruber K. Is the future of medical diagnosis in computer algorithms? Lancet Digit Health. 2019;05(1):e15\u20136.","journal-title":"Lancet Digit Health"}],"container-title":["Critical Care"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13054-023-04320-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13054-023-04320-0\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13054-023-04320-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,1,25]],"date-time":"2023-01-25T15:07:40Z","timestamp":1674659260000},"score":1,"resource":{"primary":{"URL":"https:\/\/ccforum.biomedcentral.com\/articles\/10.1186\/s13054-023-04320-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1,25]]},"references-count":28,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["4320"],"URL":"https:\/\/doi.org\/10.1186\/s13054-023-04320-0","relation":{},"ISSN":["1364-8535"],"issn-type":[{"value":"1364-8535","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1,25]]},"assertion":[{"value":"8 August 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 November 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 January 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study was conducted in accordance with the regulations of the French Commission on Information Technology and Liberties (CNIL) and was approved by the Ethics Committee of the French Intensive Care Society (#CE-SRLF-20-69). The data were anonymized and de-identified. No data were collected other than anonymized images.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable. The individual data presented in this article were anonymized and not identifiable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare that they have no competing interests","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interest"}}],"article-number":"40"}}