{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T16:46:30Z","timestamp":1792082790424,"version":"4.4.0"},"reference-count":32,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2025,5,25]],"date-time":"2025-05-25T00:00:00Z","timestamp":1748131200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>High\u2010dose\u2010rate brachytherapy (HDR\u2010BT) is an integral part of treatment for locally advanced cervical cancer, requiring accurate segmentation of the high\u2010risk clinical target volume (HR\u2010CTV) and organs at risk (OARs) on post\u2010applicator CT (pCT) for precise and safe dose delivery. Manual contouring, however, is time\u2010consuming and highly variable, with challenges heightened in cervical HDR\u2010BT due to complex anatomy and low tissue contrast. An effective auto\u2010contouring solution could significantly enhance efficiency, consistency, and accuracy in cervical HDR\u2010BT planning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop a machine learning\u2010based approach that improves the accuracy and efficiency of HR\u2010CTV and OAR segmentation on pCT images for cervical HDR\u2010BT.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      The proposed method employs two sequential deep learning models to segment target and OARs from planning CT data. The intuitive model, a U\u2010Net, initially segments simpler structures such as the bladder and HR\u2010CTV, utilizing shallow features and iodine contrast agents. Building on this, the sophisticated model targets complex structures like the sigmoid, rectum, and bowel, addressing challenges from low contrast, anatomical proximity, and imaging artifacts. This model incorporates spatial information from the intuitive model and uses total variation regularization to improve segmentation smoothness by applying a penalty to changes in gradient. This dual\u2010model approach improves accuracy and consistency in segmenting high\u2010risk clinical target volumes and organs at risk in cervical HDR\u2010BT. To validate the proposed method, 32 cervical cancer patients treated with tandem and ovoid (T&amp;O) HDR brachytherapy (3\u20135 fractions, 115 CT images) were retrospectively selected. The method's performance was assessed using four\u2010fold cross\u2010validation, comparing segmentation results to manual contours across five metrics: Dice similarity coefficient (DSC), 95% Hausdorff distance (HD\n                      <jats:sub>95<\/jats:sub>\n                      ), mean surface distance (MSD), center\u2010of\u2010mass distance (CMD), and volume difference (VD). Dosimetric evaluations included D90 for HR\u2010CTV and D2cc for OARs.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The proposed method demonstrates high segmentation accuracy for HR\u2010CTV, bladder, and rectum, achieving DSC values of 0.79\u00a0\u00b1\u00a00.06, 0.83\u00a0\u00b1\u00a00.10, and 0.76\u00a0\u00b1\u00a00.15, MSD values of 1.92\u00a0\u00b1\u00a00.77\u00a0mm, 2.24\u00a0\u00b1\u00a01.20\u00a0mm, and 4.18\u00a0\u00b1\u00a03.74\u00a0mm, and absolute VD values of 5.34\u00a0\u00b1\u00a04.85 cc, 17.16\u00a0\u00b1\u00a017.38 cc, and 18.54\u00a0\u00b1\u00a016.83 cc, respectively. Despite challenges in bowel and sigmoid segmentation due to poor soft tissue contrast in CT and variability in manual contouring (ground truth volumes of 128.48\u00a0\u00b1\u00a095.9 cc and 51.87\u00a0\u00b1\u00a040.67 cc), the method significantly outperforms two state\u2010of\u2010the\u2010art methods on DSC, MSD, and CMD metrics (\n                      <jats:italic>p<\/jats:italic>\n                      \u2010value\u00a0&lt;\u00a00.05). For HR\u2010CTV, the mean absolute D90 difference was 0.42\u00a0\u00b1\u00a01.17\u00a0Gy (\n                      <jats:italic>p<\/jats:italic>\n                      \u2010value\u00a0&gt;\u00a00.05), less than 5% of the prescription dose. Over 75% of cases showed changes within\u00a0\u00b1\u00a00.5\u00a0Gy, and fewer than 10% exceeded\u00a0\u00b1\u00a01\u00a0Gy. The mean and variation in structure volume and D2cc parameters between manual and segmented contours for OARs showed no significant differences (\n                      <jats:italic>p<\/jats:italic>\n                      \u2010value\u00a0&gt;\u00a00.05), with mean absolute D2cc differences within 0.5\u00a0Gy, except for the bladder, which exhibited higher variability (0.97\u00a0Gy).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Our innovative auto\u2010contouring method showed promising results in segmenting HR\u2010CTV and OARs from pCT, potentially enhancing the efficiency of HDR BT cervical treatment planning. Further validation and clinical implementation are required to fully realize its clinical benefits.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.17908","type":"journal-article","created":{"date-parts":[[2025,5,25]],"date-time":"2025-05-25T21:04:14Z","timestamp":1748207054000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["A novel network architecture for post\u2010applicator placement CT auto\u2010contouring in cervical cancer HDR brachytherapy"],"prefix":"10.1002","volume":"52","author":[{"given":"Yang","family":"Lei","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming","family":"Chao","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kaida","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vishal","family":"Gupta","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California at San Francisco  San Francisco California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Emi J.","family":"Yoshida","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of California at San Francisco  San Francisco California USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tingyu","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaofeng","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Winship Cancer Institute Emory University  Atlanta Georgia USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tian","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Icahn School of Medicine at Mount Sinai  New York New York USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,5,25]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1470-2045(19)30072-5"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/S2214-109X(20)30459-9"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-022-02109-2"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.6004\/jnccn.2020.0027"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.brachy.2011.07.003"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2020.04.002"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2018.11.012"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.orhc.2013.12.004"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2004.12.015"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2014.06.005"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.11.024"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.brachy.2015.09.002"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1515\/raon-2016-0023"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers15245735"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1201\/9780429323782-9"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-18084-0"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-22188-y"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.brachy.2020.03.001"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15490"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13024"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14467"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.brachy.2011.07.002"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-36940-5"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2021.03.030"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2023.101340"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2005.11.014"},{"key":"e_1_2_8_28_1","unstructured":"PanA JonesE JagadeesanM SteinhardtJ.Feedback loops with language models drive in\u2010context reward hacking.Paper presented at: Proceedings of the 41st International Conference on Machine Learning Article No. 1589 pp.39154\u201039200.2024.https:\/\/dl.acm.org\/doi\/10.5555\/3692070.3693659"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.array.2022.100258"},{"key":"e_1_2_8_30_1","doi-asserted-by":"crossref","unstructured":"LiY YuanL VasconcelosN Bidirectional learning for domain adaptation of semantic segmentation.Paper presented at: Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition. Long Beach CA USA 2019 pp.6929\u20106938 doi:10.1109\/CVPR.2019.00710","DOI":"10.1109\/CVPR.2019.00710"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12539"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.26170"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2022.100968"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.17908","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T10:34:10Z","timestamp":1765190050000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.17908"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,25]]},"references-count":32,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["10.1002\/mp.17908"],"URL":"https:\/\/doi.org\/10.1002\/mp.17908","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,25]]},"assertion":[{"value":"2024-09-10","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-05-08","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-05-25","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e17908"}}