{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T23:14:33Z","timestamp":1793229273762,"version":"4.5.0"},"reference-count":27,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T00:00:00Z","timestamp":1756944000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T00:00:00Z","timestamp":1756944000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","award":["RGPIN-2020-04741"],"award-info":[{"award-number":["RGPIN-2020-04741"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol Exp"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Lung lobe segmentation is required to assess lobar function with nuclear imaging before surgical interventions. We evaluated the performance of open-source deep learning-based lung lobe segmentation tools, compared to a similar nnU-Net model trained on a smaller but more representative clinical dataset.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>\n                      We collated and semi-automatically segmented an internal dataset of 164 computed tomography scans and classified them for task difficulty as easy, moderate, or hard. The performance of three open-source models\u2014multi-organ objective segmentation (MOOSE), TotalSegmentator, and LungMask\u2014was assessed using Dice similarity coefficient (DSC), robust Hausdorff distance (rHd95), and normalized surface distance (NSD). Additionally, we trained, validated, and tested an nnU-Net model using our local dataset and compared its performance with that of the other software on the test subset. All models were evaluated for generalizability using an external competition (LOLA11,\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200955).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      TotalSegmentator outperformed MOOSE in DSC and NSD across all difficulty levels (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001), but not in rHd95 (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20091.000). MOOSE and TotalSegmentator surpassed LungMask across metrics and difficulty classes (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001). Our model exceeded all other models on the internal dataset (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200933) in all metrics, across all difficulty classes (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001), and on the external dataset. Missing lobes were correctly identified only by our model and LungMask in 3 and 1 of 7 cases, respectively.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Open-source segmentation tools perform well in straightforward cases but struggle in unfamiliar, complex cases. Training on diverse, specialized datasets can improve generalizability, emphasizing representative data over sheer quantity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Relevance statement<\/jats:title>\n                    <jats:p>Training lung lobe segmentation models on a local variety of cases improves accuracy, thus enhancing presurgical planning, ventilation-perfusion analysis, and disease localization, potentially impacting treatment decisions and patient outcomes in respiratory and thoracic care.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:list list-type=\"bullet\">\n                        <jats:list-item>\n                          <jats:p>Deep learning models trained on non-specialized datasets struggle with complex lung anomalies, yet their real-world limitations are insufficiently assessed.<\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>Training an identical model on a smaller yet clinically diverse and representative cohort improved performance in challenging cases.<\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>Data diversity outweighs the quantity in deep learning-based segmentation models.<\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>Accurate lung lobe segmentation may enhance presurgical assessment of lung lobar ventilation and perfusion function, optimizing clinical decision-making and patient outcomes.<\/jats:p>\n                        <\/jats:list-item>\n                      <\/jats:list>\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Graphical Abstract<\/jats:title>\n                  <\/jats:sec>","DOI":"10.1186\/s41747-025-00623-9","type":"journal-article","created":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T17:29:42Z","timestamp":1757006982000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Lung lobe segmentation: performance of open-source MOOSE, TotalSegmentator, and LungMask models compared to a local in-house model"],"prefix":"10.1186","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-2259-2981","authenticated-orcid":false,"given":"Elaheh","family":"Amini","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ran","family":"Klein","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,9,4]]},"reference":[{"key":"623_CR1","doi-asserted-by":"publisher","first-page":"427","DOI":"10.1007\/s00441-016-2545-0","volume":"367","author":"JC Schittny","year":"2017","unstructured":"Schittny JC (2017) Development of the lung. Cell Tissue Res 367:427. https:\/\/doi.org\/10.1007\/s00441-016-2545-0","journal-title":"Cell Tissue Res"},{"key":"623_CR2","doi-asserted-by":"publisher","first-page":"W222","DOI":"10.2214\/AJR.12.8961","volume":"200","author":"SF Nemec","year":"2013","unstructured":"Nemec SF, Bankier AA, Eisenberg RL (2013) Upper lobe\u2013predominant diseases of the lung. AJR Am J Roentgenol 200:W222\u2013W237. https:\/\/doi.org\/10.2214\/AJR.12.8961","journal-title":"AJR Am J Roentgenol"},{"key":"623_CR3","doi-asserted-by":"publisher","unstructured":"Nemec SF, Bankier AA, Eisenberg RL et al (2013) Lower lobe\u2014predominant diseases of the lung. AJR Am J Roentgenol. https:\/\/doi.org\/10.2214\/AJR.12.9253","DOI":"10.2214\/AJR.12.9253"},{"key":"623_CR4","doi-asserted-by":"publisher","first-page":"761","DOI":"10.2214\/ajr.164.3.7863909","volume":"164","author":"RD Tarver","year":"1995","unstructured":"Tarver RD (1995) How common are incomplete pulmonary fissures, and what is their clinical significance?. AJR Am J Roentgenol 164:761. https:\/\/doi.org\/10.2214\/ajr.164.3.7863909","journal-title":"AJR Am J Roentgenol"},{"key":"623_CR5","doi-asserted-by":"publisher","first-page":"797","DOI":"10.1055\/s-0041-1741045","volume":"31","author":"M Manjunath","year":"2022","unstructured":"Manjunath M, Sharma MV, Janso K et al (2022) Study on anatomical variations in fissures of lung by CT scan. Indian J Radiol Imaging 31:797\u2013804. https:\/\/doi.org\/10.1055\/s-0041-1741045","journal-title":"Indian J Radiol Imaging"},{"key":"623_CR6","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1016\/j.compmedimag.2014.10.008","volume":"40","author":"T Doel","year":"2015","unstructured":"Doel T, Gavaghan DJ, Grau V (2015) Review of automatic pulmonary lobe segmentation methods from CT. Comput Med Imaging Graph 40:13\u201329. https:\/\/doi.org\/10.1016\/j.compmedimag.2014.10.008","journal-title":"Comput Med Imaging Graph"},{"key":"623_CR7","doi-asserted-by":"publisher","first-page":"277","DOI":"10.1055\/s-0042-1742517","volume":"31","author":"D Carmo","year":"2022","unstructured":"Carmo D, Ribeiro J, Dertkigil S et al (2022) A systematic review of automated segmentation methods and public datasets for the lung and its lobes and findings on computed tomography images. Yearb Med Inform 31:277\u2013295. https:\/\/doi.org\/10.1055\/s-0042-1742517","journal-title":"Yearb Med Inform"},{"key":"623_CR8","doi-asserted-by":"publisher","first-page":"202","DOI":"10.1109\/TMI.2008.929101","volume":"28","author":"S Ukil","year":"2009","unstructured":"Ukil S, Reinhardt JM (2009) Anatomy-guided lung lobe segmentation in X-ray CT images. IEEE Trans Med Imaging 28:202\u2013214. https:\/\/doi.org\/10.1109\/TMI.2008.929101","journal-title":"IEEE Trans Med Imaging"},{"key":"623_CR9","doi-asserted-by":"publisher","first-page":"1286","DOI":"10.1109\/TMI.2010.2044799","volume":"29","author":"EM Van Rikxoort","year":"2010","unstructured":"Van Rikxoort EM, Prokop M, De Hoop B et al (2010) Automatic segmentation of pulmonary lobes robust against incomplete fissures. IEEE Trans Med Imaging 29:1286\u20131296. https:\/\/doi.org\/10.1109\/TMI.2010.2044799","journal-title":"IEEE Trans Med Imaging"},{"key":"623_CR10","doi-asserted-by":"publisher","unstructured":"Hong Z (2020) Lung lobe segmentation based on lung fissure surface classification using a point cloud region growing approach. Algorithms. https:\/\/doi.org\/10.3390\/a13100263","DOI":"10.3390\/a13100263"},{"key":"623_CR11","doi-asserted-by":"crossref","unstructured":"Xiao C, Staring M, Wang J et al (2013) A derivative of stick filter for pulmonary fissure detection in CT images. In: Medical imaging 2013 image processing. SPIE, pp S228\u2013236","DOI":"10.1117\/12.2006566"},{"key":"623_CR12","doi-asserted-by":"publisher","first-page":"560","DOI":"10.1016\/j.compmedimag.2012.06.001","volume":"36","author":"S Gu","year":"2012","unstructured":"Gu S, Wilson D, Wang Z et al (2012) Identification of pulmonary fissures using a piecewise plane fitting algorithm. Comput Med Imaging Graph 36:560\u2013571. https:\/\/doi.org\/10.1016\/j.compmedimag.2012.06.001","journal-title":"Comput Med Imaging Graph"},{"key":"623_CR13","doi-asserted-by":"crossref","unstructured":"Ukil S, Hoffman EA, Reinhardt JM (2005) Automatic lung lobe segmentation in x-ray CT images by 3D watershed transform using anatomic information from the segmented airway tree. In: Medical imaging 2005 image Processing. SPIE, pp S556\u2013567","DOI":"10.1117\/12.596752"},{"key":"623_CR14","doi-asserted-by":"publisher","DOI":"10.1186\/1475-925X-13-59","volume":"13","author":"S Qi","year":"2014","unstructured":"Qi S, van Triest HJW, Yue Y et al (2014) Automatic pulmonary fissure detection and lobe segmentation in CT chest images. Biomed Eng Online 13: 59. https:\/\/doi.org\/10.1186\/1475-925X-13-59","journal-title":"Biomed Eng Online"},{"key":"623_CR15","doi-asserted-by":"publisher","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. 9351:234\u2013241. https:\/\/doi.org\/10.1007\/978-3-319-24574-4_28","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"623_CR16","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. https:\/\/doi.org\/10.1038\/s41592-020-01008-z","journal-title":"Nat Methods"},{"key":"623_CR17","doi-asserted-by":"crossref","unstructured":"Kikinis R, Pieper SD, Vosburgh KG (2014) 3D Slicer: a platform for subject-specific image analysis, visualization, and clinical support. In: Jolesz FA (ed) Intraoperative imaging and image-guided therapy. Springer, New York, pp S277\u2013S289","DOI":"10.1007\/978-1-4614-7657-3_19"},{"key":"623_CR18","doi-asserted-by":"publisher","first-page":"1941","DOI":"10.2967\/jnumed.122.264063","volume":"63","author":"LKS Sundar","year":"2022","unstructured":"Sundar LKS, Yu J, Muzik O et al (2022) Fully automated, semantic segmentation of whole-body 18F-FDG PET\/CT images based on data-centric artificial intelligence. J Nucl Med 63:1941\u20131948. https:\/\/doi.org\/10.2967\/jnumed.122.264063","journal-title":"J Nucl Med"},{"key":"623_CR19","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-022-01718-3","volume":"9","author":"S Gatidis","year":"2022","unstructured":"Gatidis S, Hepp T, Fr\u00fch M et al (2022) A whole-body FDG-PET\/CT dataset with manually annotated tumor lesions. Sci Data 9:601. https:\/\/doi.org\/10.1038\/s41597-022-01718-3","journal-title":"Sci Data"},{"key":"623_CR20","doi-asserted-by":"publisher","first-page":"e230024","DOI":"10.1148\/ryai.230024","volume":"5","author":"J Wasserthal","year":"2023","unstructured":"Wasserthal J, Breit H-C, Meyer MT et al (2023) TotalSegmentator: robust segmentation of 104 anatomic structures in CT images. Radiol Artif Intell 5:e230024. https:\/\/doi.org\/10.1148\/ryai.230024","journal-title":"Radiol Artif Intell"},{"key":"623_CR21","doi-asserted-by":"publisher","unstructured":"Wasserthal J (2023) Dataset with segmentations of 117 important anatomical structures in 1228 CT images. Zenodo https:\/\/doi.org\/10.5281\/zenodo.10047292","DOI":"10.5281\/zenodo.10047292"},{"key":"623_CR22","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. https:\/\/doi.org\/10.1186\/s41747-020-00173-2","journal-title":"Eur Radiol Exp"},{"key":"623_CR23","unstructured":"Lung Tissue Research Consortium (LTRC) (2019) NHLBI, NIH. 23:e26151 https:\/\/www. nhlbi.nih.gov\/science\/lung-tissue-research-consortium-ltrc"},{"key":"623_CR24","unstructured":"Reinke A, Tizabi MD, Sudre CH et al (2023) Common limitations of image processing metrics: a picture story. Preprint at https:\/\/arxiv.org\/abs\/2104.05642"},{"key":"623_CR25","doi-asserted-by":"publisher","unstructured":"Nikolov S, Blackwell S, Zverovitch A et al (2021) Clinically Applicable Segmentation of Head and Neck Anatomy for Radiotherapy: Deep Learning Algorithm Development and Validation Study. J Med Internet Res 23:e26151. https:\/\/doi.org\/10.2196\/26151","DOI":"10.2196\/26151"},{"key":"623_CR26","unstructured":"Vab Rikxoort E, van Ginneken B, Kerkstra S (2011) LOLA11\u2014grand challenge. https:\/\/lola11.grand-challenge.org\/"},{"key":"623_CR27","doi-asserted-by":"publisher","unstructured":"Paszke A, Gross S, Massa F et al (2019) PyTorch: an imperative style, high-performance deep learning library. Preprint at https:\/\/doi.org\/10.48550\/arXiv.1912.01703","DOI":"10.48550\/arXiv.1912.01703"}],"container-title":["European Radiology Experimental"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-025-00623-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s41747-025-00623-9\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-025-00623-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T17:29:45Z","timestamp":1757006985000},"score":1,"resource":{"primary":{"URL":"https:\/\/eurradiolexp.springeropen.com\/articles\/10.1186\/s41747-025-00623-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,4]]},"references-count":27,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["623"],"URL":"https:\/\/doi.org\/10.1186\/s41747-025-00623-9","relation":{},"ISSN":["2509-9280"],"issn-type":[{"value":"2509-9280","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,4]]},"assertion":[{"value":"16 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 August 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 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 database creation received approval from the Research Ethics Board of The Ottawa Hospital (Protocol number: 20220303-01H, Approval date: Jun 30, 2022), making secondary use of clinical data.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Written informed consent was not required for this study as it was a retrospective analysis of previously collected (de-identified) data.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"RK receives royalty shares from Jubilant-DraxImage from the sale of Ruby-Fill\u00ae and from Invia Medical Solutions for the sale of myocardial flow quantification software Corridor-4DM. RK has research collaboration agreements and receives research support from Jubilant-Radiopharma through MITACS and Inovait collaboration grants. RK has collaborative research agreements with Hermes Medical Solutions.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"86"}}