{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T20:35:18Z","timestamp":1783456518282,"version":"3.55.0"},"reference-count":34,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T00:00:00Z","timestamp":1759190400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T00:00:00Z","timestamp":1759190400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100002957","name":"Technische Universit\u00e4t Dresden","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100002957","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Decisions regarding veno-venous extracorporeal membrane oxygenation (vv-ECMO) in patients with acute respiratory distress syndrome (ARDS) are often based solely on clinical and physiological parameters, which may insufficiently reflect severity and heterogeneity of lung injury. This study aimed to develop a predictive model integrating machine learning-derived quantitative features from admission chest computed tomography (CT) with selected clinical variables to support early individualized decision-making regarding vv-ECMO therapy.  In this retrospective single-center cohort study, 375 consecutive patients with COVID-19-associated ARDS admitted to the ICU between March 2020 and April 2022 were included. Lung segmentation from initial CTs was performed using a convolutional neural network (CNN) to generate high-resolution, anatomically accurate masks of the lungs. Subsequently, 592 radiomic features, quantifying lung aeration, density and morphology, were extracted. Four clinical parameters \u2013 age, mean airway pressure, lactate, and C-reactive protein, were selected on the basis of clinical relevance. Three logistic regression models were developed: (1)  Imaging Model, (2) Clinical Model, and (3)  Combined Model integrating different features. Predictive performance was assessed via the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, and specificity.  A total of 375 patients were included: 172 in the training and 203 in the validation cohort. In the training cohort, the AUROCs were 0.743 (Imaging), 0.828 (Clinical), and 0.842 (Combined). In the validation cohort, the Combined Model achieved the highest AUROC (0.705), outperforming the Clinical (0.674) and Imaging (0.639) Models. Overall accuracy in the validation cohort was 64.0% (Combined), 66.5% (Clinical), and 59.1% (Imaging). The Combined Model showed 68.1% sensitivity and 58.9% specificity. Kaplan-Meier analysis confirmed a significantly greater cumulative incidence of ECMO therapy in patients predicted as high risk (\n                    <jats:italic>p<\/jats:italic>\n                    \u2009&lt;\u20090.001), underscoring its potential to support individualized, timely ECMO decisions in ARDS by providing clinicians with objective data-driven risk estimates.  Quantitative CT features based on machine learning-derived lung segmentation allow early individualized prediction of the need for vv-ECMO in ARDS. While clinical data remain essential, radiomic markers enhance prognostic accuracy. The Combined Model demonstrates considerable potential to support timely and evidence-based ECMO initiation, facilitating individualized critical care in both specialized and general ICU environments.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Trial registration<\/jats:bold>\n                    : The study is registered with the German Clinical Trials Register under the number DRKS00027856. Registered 18.01.2022, retrospectively registered due to retrospective design of the study.\n                  <\/jats:p>","DOI":"10.1038\/s41598-025-21287-w","type":"journal-article","created":{"date-parts":[[2025,9,30]],"date-time":"2025-09-30T16:55:56Z","timestamp":1759251356000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Radiomics-enhanced modelling approach for predicting the need for ECMO in ARDS patients: a retrospective cohort study"],"prefix":"10.1038","volume":"15","clinical-trial-number":[{"clinical-trial-number":"drks00027856","registry":"10.18810\/drks"}],"author":[{"given":"Martin","family":"Mirus","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eric","family":"Leitert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rebecca","family":"Bockholt","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lars","family":"Heubner","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steffen","family":"L\u00f6ck","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marie","family":"Brei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonas","family":"Biehler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jens-Peter","family":"K\u00fchn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thea","family":"Koch","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wolfgang","family":"Wall","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter Markus","family":"Spieth","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,9,30]]},"reference":[{"issue":"21","key":"21287_CR1","doi-asserted-by":"publisher","first-page":"1965","DOI":"10.1056\/NEJMoa1800385","volume":"378","author":"A Combes","year":"2018","unstructured":"Combes, A. et al. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. N. Engl. J. Med. 378 (21), 1965\u20131975 (2018).","journal-title":"N. Engl. J. Med."},{"issue":"10358","key":"21287_CR2","doi-asserted-by":"publisher","first-page":"1145","DOI":"10.1016\/S0140-6736(22)01485-4","volume":"400","author":"LDJ Bos","year":"2022","unstructured":"Bos, L. D. J. & Ware, L. B. Acute respiratory distress syndrome: causes, pathophysiology, and phenotypes. Lancet 400 (10358), 1145\u20131156 (2022).","journal-title":"Lancet"},{"issue":"10358","key":"21287_CR3","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1016\/S0140-6736(22)01439-8","volume":"400","author":"EA Gorman","year":"2022","unstructured":"Gorman, E. A., O\u2019Kane, C. M. & McAuley, D. F. Acute respiratory distress syndrome in adults: diagnosis, outcomes, long-term sequelae, and management. Lancet 400 (10358), 1157\u20131170 (2022).","journal-title":"Lancet"},{"issue":"10","key":"21287_CR4","doi-asserted-by":"publisher","first-page":"1704","DOI":"10.1007\/s00134-013-3037-2","volume":"39","author":"M Schmidt","year":"2013","unstructured":"Schmidt, M. et al. The PRESERVE mortality risk score and analysis of long-term outcomes after extracorporeal membrane oxygenation for severe acute respiratory distress syndrome. Intensive Care Med. 39 (10), 1704\u20131713 (2013).","journal-title":"Intensive Care Med."},{"issue":"2","key":"21287_CR5","doi-asserted-by":"publisher","first-page":"275","DOI":"10.1007\/s00134-012-2747-1","volume":"39","author":"F Pappalardo","year":"2013","unstructured":"Pappalardo, F. et al. Predicting mortality risk in patients undergoing venovenous ECMO for ARDS due to influenza A (H1N1) pneumonia: the ecmonet score. Intensive Care Med. 39 (2), 275\u2013281 (2013).","journal-title":"Intensive Care Med."},{"issue":"7","key":"21287_CR6","doi-asserted-by":"publisher","first-page":"1340","DOI":"10.1177\/02676591221115267","volume":"38","author":"G Majithia-Beet","year":"2023","unstructured":"Majithia-Beet, G., Naemi, R. & Issitt, R. Efficacy of outcome prediction of the respiratory ECMO survival prediction score and the predicting death for severe ARDS on VV-ECMO score for patients with acute respiratory distress syndrome on extracorporeal membrane oxygenation. Perfusion 38 (7), 1340\u20131348 (2023).","journal-title":"Perfusion"},{"key":"21287_CR7","doi-asserted-by":"crossref","unstructured":"Lara, C. A. et al. Maarten clinical decision support for ExtraCorporeal membrane oxygenation: Will we fly by wire? Perfusion 38 (1_suppl), 68\u201381. (2023).","DOI":"10.1177\/02676591231163688"},{"issue":"1","key":"21287_CR8","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1186\/s13643-023-02211-7","volume":"12","author":"L Giordano","year":"2023","unstructured":"Giordano, L. et al. Predictive models in extracorporeal membrane oxygenation (ECMO): a systematic review. Syst. Rev. 12 (1), 44 (2023).","journal-title":"Syst. Rev."},{"issue":"1","key":"21287_CR9","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1186\/s12931-023-02425-2","volume":"24","author":"M Nishikimi","year":"2023","unstructured":"Nishikimi, M. et al. Characteristics of the pulmonary opacities on chest CT associated with difficulty in short-term liberation from veno-venous ECMO in patients with severe ARDS. Respir Res. 24 (1), 128 (2023).","journal-title":"Respir Res."},{"issue":"4","key":"21287_CR10","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1016\/S2213-2600(21)00461-6","volume":"10","author":"MV Maddali","year":"2022","unstructured":"Maddali, M. V. et al. Validation and utility of ARDS subphenotypes identified by machine-learning models using clinical data: an observational, multicohort, retrospective analysis. Lancet Respiratory Med. 10 (4), 367\u2013377 (2022).","journal-title":"Lancet Respiratory Med."},{"issue":"7","key":"21287_CR11","doi-asserted-by":"publisher","first-page":"996","DOI":"10.1164\/rccm.202002-0347OC","volume":"202","author":"P Sinha","year":"2020","unstructured":"Sinha, P., Churpek, M. M. & Calfee, C. S. Machine learning classifier models can identify acute respiratory distress syndrome phenotypes using readily available clinical data. Am. J. Respir Crit. Care Med. 202 (7), 996\u20131004 (2020).","journal-title":"Am. J. Respir Crit. Care Med."},{"key":"21287_CR12","doi-asserted-by":"publisher","unstructured":"Ronneberger, O., Fischer, P. & Brox, T. U-Net: Convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells WM, Frangi AF, editors. Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015 9351 234\u201341. http:\/\/link.springer.com\/https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28 (Springer, 2015).","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"21287_CR13","doi-asserted-by":"publisher","first-page":"676118","DOI":"10.3389\/fphys.2021.676118","volume":"12","author":"P Herrmann","year":"2021","unstructured":"Herrmann, P. et al. Using artificial intelligence for automatic segmentation of CT lung images in acute respiratory distress syndrome. Front. Physiol. 12, 676118 (2021).","journal-title":"Front. Physiol."},{"key":"21287_CR14","doi-asserted-by":"publisher","unstructured":"Buvat, I. Quantitative image analysis in tomography. In: Grupen C, Buvat I, editors. Handbook of Particle Detection and Imaging 1043\u201363. https:\/\/link.springer.com\/https:\/\/doi.org\/10.1007\/978-3-642-13271-1_41  (Springer, 2012).","DOI":"10.1007\/978-3-642-13271-1_41"},{"issue":"6","key":"21287_CR15","doi-asserted-by":"publisher","first-page":"824","DOI":"10.1097\/00000542-198812000-00005","volume":"69","author":"L Gattinoni","year":"1988","unstructured":"Gattinoni, L. et al. Relationships between lung computed tomographic Density, gas Exchange, and PEEP in acute respiratory failure. Anesthesiology 69 (6), 824\u2013832 (1988).","journal-title":"Anesthesiology"},{"issue":"1","key":"21287_CR16","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1186\/s40635-014-0031-0","volume":"2","author":"A Protti","year":"2014","unstructured":"Protti, A. et al. Validation of computed tomography for measuring lung weight. ICMx 2 (1), 31 (2014).","journal-title":"ICMx"},{"key":"21287_CR17","unstructured":"Collins, G. S. et al. TRIPOD + AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ  e078378. (2024)."},{"issue":"23","key":"21287_CR18","first-page":"2526","volume":"307","author":"ARDSDT Force","year":"2012","unstructured":"Force, A. R. D. S. D. T. et al. Acute respiratory distress syndrome: the Berlin definition. JAMA 307 (23), 2526\u20132533 (2012).","journal-title":"JAMA"},{"issue":"5","key":"21287_CR19","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1097\/00004728-198010000-00017","volume":"4","author":"GN Hounsfield","year":"1980","unstructured":"Hounsfield, G. N. Computed medical imaging. J. Comput. Assist. Tomogr. 4 (5), 665\u2013674 (1980).","journal-title":"J. Comput. Assist. Tomogr."},{"key":"21287_CR20","doi-asserted-by":"crossref","unstructured":"Zwanenburg, A. & L\u00f6ck, S. familiar: end-to-end automated machine learning and model evaluation. 1.5.0.https:\/\/CRAN.R-project.org\/package=familiar (2022).","DOI":"10.32614\/CRAN.package.familiar"},{"issue":"4","key":"21287_CR21","doi-asserted-by":"publisher","first-page":"954","DOI":"10.1093\/biomet\/87.4.954","volume":"87","author":"IK Yeo","year":"2000","unstructured":"Yeo, I. K. A new family of power transformations to improve normality or symmetry. Biometrika 87 (4), 954\u2013959 (2000).","journal-title":"Biometrika"},{"issue":"8","key":"21287_CR22","doi-asserted-by":"publisher","first-page":"1226","DOI":"10.1109\/TPAMI.2005.159","volume":"27","author":"H Peng","year":"2005","unstructured":"Peng, H. & Ding, F. L. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans. Pattern Anal. Mach. Intell. 27 (8), 1226\u20131238 (2005).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"21287_CR23","doi-asserted-by":"crossref","unstructured":"Wald, R., Khoshgoftaar, T. M., Dittman, D., Awada, W. & Napolitano, A. An extensive comparison of feature ranking aggregation techniques in bioinformatics. In: 2012 IEEE 13th International Conference on Information Reuse & Integration (IRI) Las Vegas. 377\u201384.http:\/\/ieeexplore.ieee.org\/document\/6303034\/ (IEEE,  2012).","DOI":"10.1109\/IRI.2012.6303034"},{"issue":"7","key":"21287_CR24","doi-asserted-by":"publisher","first-page":"934","DOI":"10.1001\/jamainternmed.2020.0994","volume":"180","author":"C Wu","year":"2020","unstructured":"Wu, C. et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Intern. Med. 180 (7), 934 (2020).","journal-title":"JAMA Intern. Med."},{"key":"21287_CR25","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1016\/j.ajem.2023.01.032","volume":"66","author":"A Pagano","year":"2023","unstructured":"Pagano, A. et al. Blood lactate in mild and moderate ARDS secondary to SARS COV 2. Am. J. Emerg. Med. 66, 73\u201375 (2023).","journal-title":"Am. J. Emerg. Med."},{"issue":"11","key":"21287_CR26","doi-asserted-by":"publisher","first-page":"e0242917","DOI":"10.1371\/journal.pone.0242917","volume":"15","author":"T Schlesinger","year":"2020","unstructured":"Schlesinger, T. et al. (ed Di Gennaro, F.) Biodistribution and serologic response in SARS-CoV-2 induced ARDS: A cohort study. PLoS ONE 15 11 e0242917 (2020).","journal-title":"PLoS ONE"},{"issue":"4","key":"21287_CR27","doi-asserted-by":"publisher","first-page":"851","DOI":"10.1016\/j.ccc.2021.05.008","volume":"37","author":"MH Coleman","year":"2021","unstructured":"Coleman, M. H. & Aldrich, J. M. Acute respiratory distress syndrome. Crit. Care Clin. 37 (4), 851\u2013866 (2021).","journal-title":"Crit. Care Clin."},{"issue":"1","key":"21287_CR28","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, W. J. Index for rating diagnostic tests. Cancer 3 (1), 32\u201335 (1950).","journal-title":"Cancer"},{"key":"21287_CR29","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1016\/j.radonc.2018.09.007","volume":"130","author":"A Dutz","year":"2019","unstructured":"Dutz, A. & L\u00f6ck, S. Competing risks in survival data analysis. Radiother. Oncol. 130, 185\u2013189 (2019).","journal-title":"Radiother. Oncol."},{"issue":"12","key":"21287_CR30","doi-asserted-by":"publisher","first-page":"1809","DOI":"10.1097\/CCM.0000000000005673","volume":"50","author":"R Teijeiro-Paradis","year":"2022","unstructured":"Teijeiro-Paradis, R., Gannon, W. D. & Fan, E. Complications associated with venovenous extracorporeal membrane Oxygenation\u2014What can go wrong? Crit. Care Med. 50 (12), 1809\u20131818 (2022).","journal-title":"Crit. Care Med."},{"issue":"6","key":"21287_CR31","doi-asserted-by":"publisher","first-page":"1029","DOI":"10.3390\/diagnostics11061029","volume":"11","author":"E Gresser","year":"2021","unstructured":"Gresser, E. et al. Risk stratification for ECMO requirement in COVID-19 ICU patients using quantitative imaging features in CT scans on admission. Diagnostics 11 (6), 1029 (2021).","journal-title":"Diagnostics"},{"issue":"1","key":"21287_CR32","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1038\/s41746-021-00446-z","volume":"4","author":"Q Xu","year":"2021","unstructured":"Xu, Q. et al. AI-based analysis of CT images for rapid triage of COVID-19 patients. Npj Digit. Med. 4 (1), 75 (2021).","journal-title":"Npj Digit. Med."},{"issue":"1","key":"21287_CR33","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1186\/s12880-021-00640-1","volume":"21","author":"M Osadebey","year":"2021","unstructured":"Osadebey, M. et al. Three-stage segmentation of lung region from CT images using deep neural networks. BMC Med. Imaging. 21 (1), 112 (2021).","journal-title":"BMC Med. Imaging"},{"issue":"6","key":"21287_CR34","doi-asserted-by":"publisher","first-page":"717","DOI":"10.1007\/s00134-018-5064-5","volume":"44","author":"for The International ECMO Network (ECMONet) and The Extracorporeal Life Support Organization (ELSO)","year":"2018","unstructured":"for The International ECMO Network (ECMONet) and The Extracorporeal Life Support Organization (ELSO) et al. Position paper for the organization of ECMO programs for cardiac failure in adults. Intensive Care Med. 44 (6), 717\u2013729 (2018).","journal-title":"Intensive Care Med."}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-21287-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-21287-w","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-21287-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T21:01:31Z","timestamp":1764450091000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-025-21287-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,30]]},"references-count":34,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["21287"],"URL":"https:\/\/doi.org\/10.1038\/s41598-025-21287-w","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,30]]},"assertion":[{"value":"26 May 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 September 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 September 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"The study was conducted in accordance with the Declaration of Helsinki and approved by the responsible Ethics Committee of the Technical University of Dresden, Germany (BO-EK-374072021).","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}}],"article-number":"34120"}}