{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T14:59:59Z","timestamp":1787237999319,"version":"build-2736575974"},"reference-count":26,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2025,4,5]],"date-time":"2025-04-05T00:00:00Z","timestamp":1743811200000},"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":["J Applied Clin Med Phys"],"published-print":{"date-parts":[[2025,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>The use of deep learning\u2010based auto\u2010contouring algorithms in various treatment planning services is increasingly common. There is a notable deficit of commercially or publicly available models trained on large or diverse datasets containing high\u2010dose\u2010rate (HDR) brachytherapy treatment scans, leading to poor performance on images that include HDR implants.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To implement and evaluate automatic organs\u2010at\u2010risk (OARs) segmentation models for use in prostatic\u2010and\u2010gynecological computed tomography (CT)\u2010guided high\u2010dose\u2010rate brachytherapy treatment planning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods and materials<\/jats:title>\n                    <jats:p>1316 computed tomography (CT) scans and corresponding segmentation files from 1105 prostatic\u2010or\u2010gynecological HDR patients treated at our institution from 2017 to 2024 were used for model training. Data sources comprised six CT scanners including a mobile CT unit with previously reported susceptibility to image streaking artifacts. Two UNet\u2010derived model architectures, UNet++ and nnU\u2010Net, were investigated for bladder and rectum model training. The models were tested on 100 CT scans and clinically\u2010used segmentation files from 62 prostatic\u2010or\u2010gynecological HDR brachytherapy patients, disjoint from the training set, collected in 2024. Performance was evaluated using the Dice\u2010Similarity\u2010Coefficient (DSC) between model predicted contours and clinically\u2010used contours on slices in common with the Clinical\u2010Target\u2010Volume (CTV). Additionally, a blinded evaluation of ten random test cases was conducted by three experienced planners.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Median (interquartile range) 3D DSC on CTV\u2010containing slices were 0.95 (0.04) and 0.87 (0.09) for the UNet++ bladder and rectum models, respectively, and 0.96 (0.03) and 0.88 (0.10) for the nnU\u2010Net. The rank\u2010sum test did not reveal statistically significant differences in these DSC (\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0=\u00a00.15 and 0.27, respectively). The blinded evaluation scored trained models higher than clinically\u2010used contours.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Both UNet\u2010derived architectures perform similarly on the bladder and rectum and are adequately accurate to reduce contouring time in a review\u2010and\u2010edit context during HDR brachytherapy planning. The UNet++ models were chosen for implementation at our institution due to lower computing hardware requirements and are in routine clinical use.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/acm2.70089","type":"journal-article","created":{"date-parts":[[2025,4,5]],"date-time":"2025-04-05T21:55:42Z","timestamp":1743890142000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Open\u2010source deep\u2010learning models for segmentation of normal structures for prostatic and gynecological high\u2010dose\u2010rate brachytherapy: Comparison of architectures"],"prefix":"10.1002","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7524-5569","authenticated-orcid":false,"given":"Andrew J.","family":"Krupien","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yasin","family":"Abdulkadir","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dishane C.","family":"Luximon","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7907-8748","authenticated-orcid":false,"given":"John","family":"Charters","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2695-4251","authenticated-orcid":false,"given":"Huiming","family":"Dong","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan","family":"Pham","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0752-5825","authenticated-orcid":false,"given":"Dylan","family":"O'Connell","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3421-7481","authenticated-orcid":false,"given":"Jack","family":"Neylon","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James M.","family":"Lamb","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California  Los Angeles California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,4,5]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13040667"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.canrad.2021.06.023"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers15245735"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16676"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2024.02.041"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13988"},{"key":"e_1_2_10_8_1","volume-title":"StatPearls","author":"Mayer C","year":"2023"},{"key":"e_1_2_10_9_1","doi-asserted-by":"publisher","DOI":"10.2147\/CMAR.S46042"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13024"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010022\u201002121\u20103"},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.14371"},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/acba76"},{"key":"e_1_2_10_14_1","unstructured":"RonnebergerO FischerP BroxT.U\u2010Net: convolutional networks for biomedical image segmentation. arXiv.org. Published May 18 2015.https:\/\/arxiv.org\/abs\/1505.04597"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592\u2010020\u201001008\u2010z"},{"key":"e_1_2_10_16_1","unstructured":"ZhouZ RahmanM TajbakhshN LiangJ.UNet++: a nested U\u2010Net architecture for medical image segmentation. arXiv.org.2018.https:\/\/arxiv.org\/abs\/1807.10165"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010024\u201044824\u2010z"},{"key":"e_1_2_10_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12084"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12738"},{"key":"e_1_2_10_20_1","doi-asserted-by":"crossref","volume-title":"Preprocessing for image classification by convolutional neural networks","author":"Pal KK","DOI":"10.1109\/RTEICT.2016.7808140"},{"key":"e_1_2_10_21_1","doi-asserted-by":"publisher","DOI":"10.3390\/app131910760"},{"key":"e_1_2_10_22_1","doi-asserted-by":"crossref","volume-title":"A survey of loss functions for semantic segmentation","author":"Jadon S","DOI":"10.1109\/CIBCB48159.2020.9277638"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13568"},{"key":"e_1_2_10_24_1","unstructured":"KirillovA MintunE RaviN et\u00a0al.Segment anything. arXiv:230402643 [cs]. Published online April 5 2023.https:\/\/arxiv.org\/abs\/2304.02643"},{"key":"e_1_2_10_25_1","unstructured":"HeS BaoR LiJ GrantPE OuY.Accuracy of segment\u2010anything model (SAM) in medical image segmentation tasks. arXiv:230409324 [cs eess]. Published online April 27 2023.https:\/\/arxiv.org\/abs\/2304.09324"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.230024"},{"key":"e_1_2_10_27_1","unstructured":"RaviN GabeurV HuYT et\u00a0al.SAM 2: segment anything in images and videos. arXiv.org.2024.https:\/\/arxiv.org\/abs\/2408.00714"}],"container-title":["Journal of Applied Clinical Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/acm2.70089","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,18]],"date-time":"2025-09-18T20:54:48Z","timestamp":1758228888000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/acm2.70089"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,5]]},"references-count":26,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2025,6]]}},"alternative-id":["10.1002\/acm2.70089"],"URL":"https:\/\/doi.org\/10.1002\/acm2.70089","archive":["Portico"],"relation":{},"ISSN":["1526-9914","1526-9914"],"issn-type":[{"value":"1526-9914","type":"print"},{"value":"1526-9914","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,5]]},"assertion":[{"value":"2024-10-22","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-16","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-05","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70089"}}