{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T19:38:56Z","timestamp":1792697936197,"version":"4.5.0"},"reference-count":27,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2023,9,29]],"date-time":"2023-09-29T00:00:00Z","timestamp":1695945600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,9,29]],"date-time":"2023-09-29T00:00:00Z","timestamp":1695945600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100010269","name":"Wellcome Trust","doi-asserted-by":"publisher","award":["222930\/Z\/21\/Z"],"award-info":[{"award-number":["222930\/Z\/21\/Z"]}],"id":[{"id":"10.13039\/100010269","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100010269","name":"Wellcome Trust","doi-asserted-by":"publisher","award":["4ward North 203914\/Z\/16 for Krit Dwivedi (K.D.). Andrew Swift AJS 205188\/Z\/16\/Z"],"award-info":[{"award-number":["4ward North 203914\/Z\/16 for Krit Dwivedi (K.D.). Andrew Swift AJS 205188\/Z\/16\/Z"]}],"id":[{"id":"10.13039\/100010269","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol"],"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Objectives<\/jats:title>\n                <jats:p>There is a need for CT pulmonary angiography (CTPA) lung segmentation models. Clinical translation requires radiological evaluation of model outputs, understanding of limitations, and identification of failure points. This multicentre study aims to develop an accurate CTPA lung segmentation model, with evaluation of outputs in two diverse patient cohorts with pulmonary hypertension (PH) and interstitial lung disease (ILD).<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>This retrospective study develops an nnU-Net-based segmentation model using data from two specialist centres (UK and USA). Model was trained (<jats:italic>n<\/jats:italic>\u2009=\u200937), tested (<jats:italic>n<\/jats:italic>\u2009=\u200912), and clinically evaluated (<jats:italic>n<\/jats:italic>\u2009=\u2009176) on a diverse \u2018real-world\u2019 cohort of 225 PH patients with volumetric CTPAs. Dice score coefficient (DSC) and normalised surface distance (NSD) were used for testing. Clinical evaluation of outputs was performed by two radiologists who assessed clinical significance of errors. External validation was performed on heterogenous contrast and non-contrast scans from 28 ILD patients.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>A total of 225 PH and 28 ILD patients with diverse demographic and clinical characteristics were evaluated. Mean accuracy, DSC, and NSD scores were 0.998 (95% CI 0.9976, 0.9989), 0.990 (0.9840, 0.9962), and 0.983 (0.9686, 0.9972) respectively. There were no segmentation failures. On radiological review, 82% and 71% of internal and external cases respectively had no errors. Eighteen percent and 25% respectively had clinically insignificant errors. Peripheral atelectasis and consolidation were common causes for suboptimal segmentation. One external case (0.5%) with patulous oesophagus had a clinically significant error.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusion<\/jats:title>\n                <jats:p>State-of-the-art CTPA lung segmentation model provides accurate outputs with minimal clinical errors on evaluation across two diverse cohorts with PH and ILD.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Clinical relevance<\/jats:title>\n                <jats:p>Clinical translation of artificial intelligence models requires radiological review and understanding of model limitations. This study develops an externally validated state-of-the-art model with robust radiological review. Intended clinical use is in techniques such as lung volume or parenchymal disease quantification.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Key Points<\/jats:title>\n                <jats:p>\u2022 <jats:italic>Accurate, externally validated CT pulmonary angiography (CTPA) lung segmentation model tested in two large heterogeneous clinical cohorts (pulmonary hypertension and interstitial lung disease).<\/jats:italic><\/jats:p>\n                <jats:p>\u2022 <jats:italic>No segmentation failures and robust review of model outputs by radiologists found 1 (0.5%) clinically significant segmentation error.<\/jats:italic><\/jats:p>\n                <jats:p>\u2022<jats:italic> Intended clinical use of this model is a necessary step in techniques such as lung volume, parenchymal disease quantification, or pulmonary vessel analysis.<\/jats:italic><\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Graphical Abstract<\/jats:title>\n                \n              <\/jats:sec>","DOI":"10.1007\/s00330-023-10235-9","type":"journal-article","created":{"date-parts":[[2023,9,29]],"date-time":"2023-09-29T19:01:37Z","timestamp":1696014097000},"page":"2727-2737","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":14,"title":["External validation, radiological evaluation, and development of deep learning automatic lung segmentation in contrast-enhanced chest CT"],"prefix":"10.1007","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4949-4550","authenticated-orcid":false,"given":"Krit","family":"Dwivedi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Sharkey","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samer","family":"Alabed","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Curtis P.","family":"Langlotz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andy J.","family":"Swift","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christian","family":"Bluethgen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,9,29]]},"reference":[{"key":"10235_CR1","doi-asserted-by":"publisher","first-page":"582","DOI":"10.1007\/s10278-019-00227-x","volume":"32","author":"MH Hesamian","year":"2019","unstructured":"Hesamian MH, Jia W, He X, Kennedy P (2019) Deep learning techniques for medical image segmentation: achievements and challenges. J Digit Imaging 32:582\u2013596","journal-title":"J Digit Imaging"},{"key":"10235_CR2","doi-asserted-by":"publisher","first-page":"1695","DOI":"10.1016\/j.acra.2019.07.006","volume":"26","author":"L Lenchik","year":"2019","unstructured":"Lenchik L, Heacock L, Weaver AA et al (2019) Automated segmentation of tissues using CT and MRI: a systematic review. Acad Radiol 26:1695\u20131706","journal-title":"Acad Radiol"},{"key":"10235_CR3","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1186\/s41747-020-00173-2","volume":"4","author":"J Hofmanninger","year":"2020","unstructured":"Hofmanninger J, Prayer F, Pan J et al (2020) Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem. Eur Radiol Exp 4:50","journal-title":"Eur Radiol Exp"},{"key":"10235_CR4","doi-asserted-by":"publisher","first-page":"154007","DOI":"10.1109\/ACCESS.2020.3018666","volume":"8","author":"W Cao","year":"2020","unstructured":"Cao W, Wu R, Cao G, He Z (2020) A comprehensive review of computer-aided diagnosis of pulmonary nodules based on computed tomography scans. IEEE Access 8:154007\u2013154023","journal-title":"IEEE Access"},{"key":"10235_CR5","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1007\/s10278-018-0101-z","volume":"31","author":"LE Carvalho","year":"2018","unstructured":"Carvalho LE, Sobieranski AC, von Wangenheim A (2018) 3D segmentation algorithms for computerized tomographic imaging: a systematic literature review. J Digit Imaging 31:799\u2013850","journal-title":"J Digit Imaging"},{"key":"10235_CR6","doi-asserted-by":"publisher","unstructured":"Gerard SE, Herrmann J, Xin Y, et al (2020) CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network. arXiv:2010.08582 [eess.IV] https:\/\/doi.org\/10.48550\/arXiv.2010.08582","DOI":"10.48550\/arXiv.2010.08582"},{"key":"10235_CR7","first-page":"1959","volume":"12","author":"J Sousa","year":"2022","unstructured":"Sousa J, Pereira T, Silva F et al (2022) Lung segmentation in CT images: a residual U-Net approach on a cross-cohort dataset. NATO Adv Sci Inst Ser E Appl Sci 12:1959","journal-title":"NATO Adv Sci Inst Ser E Appl Sci"},{"key":"10235_CR8","doi-asserted-by":"publisher","first-page":"476","DOI":"10.3348\/kjr.2020.0318","volume":"22","author":"SJ Yoo","year":"2021","unstructured":"Yoo SJ, Yoon SH, Lee JH et al (2021) Automated lung segmentation on chest computed tomography images with extensive lung parenchymal abnormalities using a deep neural network. Korean J Radiol 22:476\u2013488","journal-title":"Korean J Radiol"},{"key":"10235_CR9","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","volume":"18","author":"F Isensee","year":"2021","unstructured":"Isensee F, Jaeger PF, Kohl SAA et al (2021) nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods 18:203\u2013211","journal-title":"Nat Methods"},{"key":"10235_CR10","doi-asserted-by":"publisher","first-page":"945","DOI":"10.1183\/09031936.00078411","volume":"39","author":"J Hurdman","year":"2012","unstructured":"Hurdman J, Condliffe R, Elliot CA et al (2012) ASPIRE registry: Assessing the Spectrum of Pulmonary hypertension Identified at a REferral centre. Eur Respir J 39:945\u2013955","journal-title":"Eur Respir J"},{"key":"10235_CR11","doi-asserted-by":"publisher","first-page":"2000041","DOI":"10.1183\/13993003.00041-2020","volume":"55","author":"RA Lewis","year":"2020","unstructured":"Lewis RA, Thompson AAR, Billings CG et al (2020) Mild parenchymal lung disease and\/or low diffusion capacity impacts survival and treatment response in patients diagnosed with idiopathic pulmonary arterial hypertension. Eur Respir J 55:2000041. https:\/\/doi.org\/10.1183\/13993003.00041-2020","journal-title":"Eur Respir J"},{"key":"10235_CR12","doi-asserted-by":"publisher","first-page":"e155","DOI":"10.1097\/as9.0000000000000155","volume":"3","author":"RW Pettit","year":"2022","unstructured":"Pettit RW, Marlatt BB, Corr SJ et al (2022) nnU-Net deep learning method for segmenting parenchyma and determining liver volume from computed tomography images. Ann Surg Open 3:e155. https:\/\/doi.org\/10.1097\/as9.0000000000000155","journal-title":"Ann Surg Open"},{"key":"10235_CR13","doi-asserted-by":"publisher","unstructured":"Reinke A, Eisenmann M, Tizabi MD, et al (2022) Common limitations of performance metrics in biomedical image analysis arXiv:2104.05642 [eess.IV] https:\/\/doi.org\/10.48550\/arXiv.2104.05642","DOI":"10.48550\/arXiv.2104.05642"},{"key":"10235_CR14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.cpet.2021.09.007","volume":"17","author":"N Hasani","year":"2022","unstructured":"Hasani N, Morris MA, Rhamim A et al (2022) Trustworthy artificial intelligence in medical imaging. PET Clin 17:1\u201312","journal-title":"PET Clin"},{"key":"10235_CR15","unstructured":"RCR position statement on artificial intelligence. https:\/\/www.rcr.ac.uk\/posts\/rcr-position-statement-artificial-intelligence. Accessed 11 Dec 2020"},{"key":"10235_CR16","doi-asserted-by":"publisher","first-page":"405","DOI":"10.3348\/kjr.2019.0025","volume":"20","author":"DW Kim","year":"2019","unstructured":"Kim DW, Jang HY, Kim KW et al (2019) Design characteristics of studies reporting the performance of artificial intelligence algorithms for diagnostic analysis of medical images: results from recently published papers. Korean J Radiol 20:405\u2013410","journal-title":"Korean J Radiol"},{"key":"10235_CR17","doi-asserted-by":"publisher","first-page":"e210064","DOI":"10.1148\/ryai.210064","volume":"4","author":"AC Yu","year":"2022","unstructured":"Yu AC, Mohajer B, Eng J (2022) External validation of deep learning algorithms for radiologic diagnosis: a systematic review. Radiol Artif Intell 4:e210064","journal-title":"Radiol Artif Intell"},{"key":"10235_CR18","doi-asserted-by":"publisher","first-page":"e210095","DOI":"10.1148\/ryai.210095","volume":"4","author":"F Sforazzini","year":"2022","unstructured":"Sforazzini F, Salome P, Moustafa M et al (2022) Deep learning-based automatic lung segmentation on multiresolution CT scans from healthy and fibrotic lungs in mice. Radiol Artif Intell 4:e210095","journal-title":"Radiol Artif Intell"},{"key":"10235_CR19","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1007\/s10278-019-00223-1","volume":"33","author":"J Park","year":"2020","unstructured":"Park J, Yun J, Kim N et al (2020) Fully automated lung lobe segmentation in volumetric chest CT with 3D U-Net: validation with intra- and extra-datasets. J Digit Imaging 33:221\u2013230","journal-title":"J Digit Imaging"},{"key":"10235_CR20","doi-asserted-by":"publisher","unstructured":"Roth H, Xu Z, Diez CT, et al (2021) Rapid artificial intelligence solutions in a pandemic - the COVID-19\u201320 lung CT lesion segmentation challenge. Res Sq. https:\/\/doi.org\/10.21203\/rs.3.rs-571332\/v1","DOI":"10.21203\/rs.3.rs-571332\/v1"},{"key":"10235_CR21","doi-asserted-by":"publisher","first-page":"4568","DOI":"10.1002\/mp.13141","volume":"45","author":"J Yang","year":"2018","unstructured":"Yang J, Veeraraghavan H, Armato SG 3rd et al (2018) Autosegmentation for thoracic radiation treatment planning: a grand challenge at AAPM 2017. Med Phys 45:4568\u20134581","journal-title":"Med Phys"},{"key":"10235_CR22","doi-asserted-by":"publisher","unstructured":"Astley JR, Biancardi AM, Hughes PJC, et al (2023) Implementable deep learning for multi-sequence proton MRI lung segmentation: a multi-center, multi-vendor, and multi-disease study. J Magn Reson Imaging.https:\/\/doi.org\/10.1002\/jmri.28643","DOI":"10.1002\/jmri.28643"},{"key":"10235_CR23","doi-asserted-by":"publisher","first-page":"204589402199996","DOI":"10.1177\/2045894021999960","volume":"11","author":"SD Nathan","year":"2021","unstructured":"Nathan SD, Barnett SD, King CS et al (2021) Impact of the new definition for pulmonary hypertension in patients with lung disease: an analysis of the United Network for Organ Sharing database. Pulm Circ 11:2045894021999960","journal-title":"Pulm Circ"},{"key":"10235_CR24","doi-asserted-by":"publisher","unstructured":"Dwivedi K, Condliffe R, Sharkey M, et al (2022) Computed tomography lung parenchymal descriptions in routine radiological reporting have diagnostic and prognostic utility in patients with idiopathic pulmonary arterial hypertension and pulmonary hypertension associated with lung disease. ERJ Open Res 8:. https:\/\/doi.org\/10.1183\/23120541.00549-2021","DOI":"10.1183\/23120541.00549-2021"},{"key":"10235_CR25","doi-asserted-by":"publisher","first-page":"e90","DOI":"10.1016\/S2213-2600(22)00273-9","volume":"10","author":"MM Hoeper","year":"2022","unstructured":"Hoeper MM, Vonk-Noordegraaf A, Kiely DG (2022) Phenotypes of idiopathic pulmonary arterial hypertension - authors\u2019 reply. Lancet Respir Med 10:e90\u2013e91","journal-title":"Lancet Respir Med"},{"key":"10235_CR26","doi-asserted-by":"publisher","first-page":"957","DOI":"10.1148\/rg.2019180168","volume":"39","author":"FU Kay","year":"2019","unstructured":"Kay FU, Oz OK, Abbara S et al (2019) Translation of quantitative imaging biomarkers into clinical chest CT. Radiographics 39:957\u2013976","journal-title":"Radiographics"},{"key":"10235_CR27","doi-asserted-by":"publisher","unstructured":"Gopalan D, Gibbs JSR (2020) From early morphometrics to machine learning-what future for cardiovascular imaging of the pulmonary circulation? Diagnostics (Basel) 10:. https:\/\/doi.org\/10.3390\/diagnostics10121004","DOI":"10.3390\/diagnostics10121004"}],"container-title":["European Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-023-10235-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00330-023-10235-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-023-10235-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,21]],"date-time":"2024-03-21T09:08:55Z","timestamp":1711012135000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00330-023-10235-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,9,29]]},"references-count":27,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,4]]}},"alternative-id":["10235"],"URL":"https:\/\/doi.org\/10.1007\/s00330-023-10235-9","relation":{},"ISSN":["1432-1084"],"issn-type":[{"value":"1432-1084","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,9,29]]},"assertion":[{"value":"28 April 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 June 2023","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 July 2023","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 September 2023","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The scientific guarantor of this publication is Andrew J. Swift.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Guarantor"}},{"value":"The authors of this manuscript declare relationships with the following companies:Conflicts of interest disclosures are as follows. None are directly related to this submitted work but have been extensively detailed for full disclosure.K. Dwivedi reports funding from Janssen Pharmaceuticals, outside the submitted work, and funding from the National Institute of Health Research UK.A. Swift reports honoraria and consultancy fees from Janssen Pharmaceuticals and received a research grant from GlaxoSmithKline. Christian Bluethgen reports support from the Kurt and Senta Herrmann Foundation and the Swiss Society of Radiology.M.Sharkey and S. Alabed have received funding from the National Institute of Health Research UK.Curtis Langlotz reports personal financial interests:Board of directors and shareholder, Bunkerhill Health.Option holder, whiterabbit.ai.Advisor and option holder, GalileoCDS.Advisor and option holder, Sirona Medical.Advisor and option holder, Adra.Advisor and option holder, Kheiron.Advisor, Sixth Street.Recent grant and gift support paid to institution:BunkerHill Health; Carestream; CARPL; Clarity; GE Healthcare; Google Cloud; IBM; IDEXX; Hospital Israelita Albert Einstein; Kheiron; Lambda; Lunit; Microsoft; Nightingale Open Science; Nines; Philips; Siemens Healthineers; Subtle Medical; VinBrain; Whiterabbit.ai; Lowenstein Foundation; Gordon and Betty Moore Foundation; Paustenbach Fund","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Multiple authors have significant statistical expertise. No complex statistical methods were necessary for this paper.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Statistics and biometry"}},{"value":"Written informed consent was waived by the Institutional Review Board.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed consent"}},{"value":"Institutional Review Board approval was obtained. Ethical approval was granted by the Institutional Review Board at both centres and approved by the UK National Research Ethics Service (16\/YH\/0352).","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"There may be subject overlap at the patient level with previously published works, as there have been multiple publications that have used the registry data that this study is based upon, the ASPIRE (Assessing the Severity of Pulmonary Hypertension In a Pulmonary Hypertension REferral Centre) registry.","order":7,"name":"Ethics","group":{"name":"EthicsHeading","label":"Study subjects or cohorts overlap"}},{"value":"\u2022 prospective\u2022 experimental\u2022 multicentre study","order":8,"name":"Ethics","group":{"name":"EthicsHeading","label":"Methodology"}}]}}