{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T21:45:58Z","timestamp":1783979158565,"version":"3.55.0"},"reference-count":37,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>As auto\u2010segmentation tools become integral to radiotherapy, more commercial products emerge. However, they may not always suit our needs. One notable example is the use of adult\u2010trained commercial software for the contouring of organs at risk (OARs) of pediatric patients.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aimed to compare three auto\u2010segmentation approaches in the context of pediatric craniospinal irradiation (CSI): commercial, out\u2010of\u2010the\u2010box, and in\u2010house.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>CT scans from 142 pediatric patients undergoing CSI were obtained from St. Jude Children's Research Hospital (training: 115; validation: 27). A test dataset comprising 16 CT scans was collected from the McGill University Health Centre. All images underwent manual delineation of 18 OARs. LimbusAI v1.7 served as the commercial product, while nnU\u2010Net was trained for benchmarking. Additionally, a two\u2010step in\u2010house approach was pursued where smaller 3D CT scans containing the OAR of interest were first recovered and then used as input to train organ\u2010specific models. Three variants of the U\u2010Net architecture were explored: a basic U\u2010Net, an attention U\u2010Net, and a 2.5D U\u2010Net. The dice similarity coefficient (DSC) assessed segmentation accuracy, and the DSC trend with age was investigated (Mann\u2013Kendall test). A radiation oncologist determined the clinical acceptability of all contours using a five\u2010point Likert scale.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Differences in the contours between the validation and test datasets reflected the distinct institutional standards. The lungs and left kidney displayed an increasing age\u2010related trend of the DSC values with LimbusAI on the validation and test datasets. LimbusAI contours of the esophagus were often truncated distally and mistaken for the trachea for younger patients, resulting in a DSC score of less than 0.5 on both datasets. Additionally, the kidneys frequently exhibited false negatives, leading to mean DSC values that were up to 0.11 lower on the validation set and 0.07 on the test set compared to the other models. Overall, nnU\u2010Net achieved good performance for body organs but exhibited difficulty differentiating the laterality of head structures, resulting in a large variation of DSC values with the standard deviation reaching 0.35 for the lenses. All in\u2010house models generally had similar DSC values when compared against each other and nnU\u2010Net. Inference time on the test data was between 47\u201355\u00a0min on a Central Processing Unit (CPU) for the in\u2010house models, while it was 1h\u00a021m with a V100 Graphics Processing Unit (GPU) for nnU\u2010Net.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>LimbusAI could not adapt well to pediatric anatomy for the esophagus and the kidneys. When commercial products do not suit the study population, the nnU\u2010Net is a viable option but requires adjustments. In resource\u2010constrained settings, the in\u2010house model provides an alternative. Implementing an automated segmentation tool requires careful monitoring and quality assurance regardless of the approach.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17782","type":"journal-article","created":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T00:00:18Z","timestamp":1743724818000},"page":"3541-3556","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Trade\u2010off of different deep learning\u2010based auto\u2010segmentation approaches for treatment planning of pediatric craniospinal irradiation autocontouring of OARs for pediatric CSI"],"prefix":"10.1002","volume":"52","author":[{"given":"Alana","family":"Thibodeau\u2010Antonacci","sequence":"first","affiliation":[{"name":"Medical Physics Unit Department of Oncology McGill University  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marija","family":"Popovic","sequence":"additional","affiliation":[{"name":"Medical Physics Unit Department of Oncology McGill University  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ozgur","family":"Ates","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology St. Jude Children's Research Hospital  Memphis USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chia\u2010Ho","family":"Hua","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology St. Jude Children's Research Hospital  Memphis USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James","family":"Schneider","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Jewish General Hospital  Montreal Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sonia","family":"Skamene","sequence":"additional","affiliation":[{"name":"Gerald Bronfman Department of Oncology McGill University  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carolyn","family":"Freeman","sequence":"additional","affiliation":[{"name":"Gerald Bronfman Department of Oncology McGill University  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shirin Abbasinejad","family":"Enger","sequence":"additional","affiliation":[{"name":"Medical Physics Unit Department of Oncology McGill University  Montreal Quebec Canada"},{"name":"Lady Davis Institute for Medical Research Jewish General Hospital  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James Man Git","family":"Tsui","sequence":"additional","affiliation":[{"name":"Gerald Bronfman Department of Oncology McGill University  Montreal Quebec Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,4]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1111\/1754\u20109485.12462"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2012.08.008"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14320"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010021\u201002330\u2010y"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.41260"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010021\u201001896\u20101"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2023.101177"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.3390\/life12122088"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2019.10.019"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2021.626499"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.3390\/ijerph19159057"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010021\u201001831\u20104"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2020.05.013"},{"key":"e_1_2_8_15_1","unstructured":"LimbusAI Press\u2014The latest news from Limbus AI. Accessed December 14 2024.https:\/\/limbus.ai\/press"},{"key":"e_1_2_8_16_1","unstructured":"IsenseeF PetersenJ KleinA et\u00a0al.nnU\u2010Net: Self\u2010adapting Framework for U\u2010Net\u2010Based Medical Image Segmentation.arXiv;2018. Accessed February 4 2024.http:\/\/arxiv.org\/abs\/1809.10486"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592\u2010020\u201001008\u2010z"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/pbc.30164"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmir.2020.11.013"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010021\u201084698\u20105"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2020.11.037"},{"key":"e_1_2_8_22_1","doi-asserted-by":"crossref","unstructured":"RedmonJ DivvalaS GirshickR FarhadiA You only look once: unified real\u2010time object detection. In:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE;2016:779\u2010788. doi:10.1109\/CVPR.2016.91","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_24_1","unstructured":"OktayO SchlemperJ FolgocLL et\u00a0al.Attention U\u2010Net: learning where to look for the pancreas. In:Computer Vision and Pattern Recognition. arXiv;2018. Accessed February 4 2024 http:\/\/arxiv.org\/abs\/1804.03999"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2021.04.019"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13040667"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.21105\/joss.01556"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2024.01.201"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010024\u201055015\u20107"},{"key":"e_1_2_8_30_1","article-title":"Common limitations of performance metrics in biomedical image analysis","author":"Reinke A","journal-title":"Proc Med Imaging Deep Learn"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2023.10.011"},{"key":"e_1_2_8_32_1","doi-asserted-by":"crossref","unstructured":"GirshickR DonahueJ DarrellT MalikJ.Rich feature hierarchies for accurate object detection and semantic segmentation.In 2014 IEEE Conference on Computer Vision and Pattern Recognition 580-87. Columbus OH USA: IEEE. doi:10.1109\/CVPR.2014.81","DOI":"10.1109\/CVPR.2014.81"},{"key":"e_1_2_8_33_1","first-page":"213","volume-title":"Computer Vision\u2014ECCV 2020. Lecture Notes in Computer Science","author":"Carion N","year":"2020"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256\u2010019\u20100099\u2010z"},{"key":"e_1_2_8_35_1","first-page":"158","volume-title":"Machine Learning and Medical Engineering for Cardiovascular Health and Intravascular Imaging and Computer Assisted Stenting. Lecture Notes in Computer Science","author":"Angermann C","year":"2019"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2023.100515"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2023.10.033"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2020.3003240"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17782","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,12]],"date-time":"2025-11-12T08:03:45Z","timestamp":1762934625000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17782"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4]]},"references-count":37,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,6]]}},"alternative-id":["10.1002\/mp.17782"],"URL":"https:\/\/doi.org\/10.1002\/mp.17782","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4]]},"assertion":[{"value":"2024-06-28","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-13","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-01","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}