{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,14]],"date-time":"2026-10-14T20:31:57Z","timestamp":1792009917801,"version":"4.4.0"},"reference-count":18,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2021,5,14]],"date-time":"2021-05-14T00:00:00Z","timestamp":1620950400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/100000054","name":"National Cancer Institute","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000054","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2021,6]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Accurate deformable registration between computed tomography (CT) and cone\u2010beam CT (CBCT) images of pancreatic cancer patients treated with high biologically effective radiation doses is essential to assess changes in organ\u2010at\u2010risk (OAR) locations and shapes and to compute delivered dose. This study describes the development and evaluation of a deep\u2010learning (DL) registration model to predict OAR segmentations on the CBCT derived from segmentations on the planning CT.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      The DL model is trained with CT\u2010CBCT image pairs of the same patient, on which OAR segmentations of the small bowel, stomach, and duodenum have been manually drawn. A transformation map is obtained, which serves to warp the CT image and segmentations. In addition to a regularity loss and an image similarity loss, an OAR segmentation similarity loss is also used during training, which penalizes the mismatch between warped CT segmentations and manually drawn CBCT segmentations. At test time, CBCT segmentations are not required as they are instead obtained from the warped CT segmentations. In an IRB\u2010approved retrospective study, a dataset consisting of 40 patients, each with one planning CT and two CBCT scans, was used in a fivefold cross\u2010validation to train and evaluate the model, using physician\u2010drawn segmentations as reference. Images were preprocessed to remove gas pockets. Network performance was compared to two intensity\u2010based deformable registration algorithms (large deformation diffeomorphic metric mapping [LDDMM] and multimodality free\u2010form [MMFF]) as baseline. Evaluated metrics were Dice similarity coefficient (DSC), change in OAR volume within a volume of interest (enclosing the low\u2010dose PTV plus 1\u00a0cm margin) from planning CT to CBCT, and maximum dose to 5 cm\n                      <jats:sup>3<\/jats:sup>\n                      of the OAR [D(5cc)].\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Processing time for one CT\u2010CBCT registration with the DL model at test time was less than 5\u00a0seconds on a GPU\u2010based system, compared to an average of 30\u00a0minutes for LDDMM optimization. For both small bowel and stomach\/duodenum, the DL model yielded larger median DSC and smaller interquartile variation than either MMFF (paired t\u2010test\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a010\n                      <jats:sup>\u22124<\/jats:sup>\n                      for both type of OARs) or LDDMM (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a010\n                      <jats:sup>\u22123<\/jats:sup>\n                      and\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0=\u00a00.03 respectively). Root\u2010mean\u2010square deviation (RMSD) of DL\u2010predicted change in small bowel volume relative to reference was 22% less than for MMFF (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0=\u00a00.007). RMSD of DL\u2010predicted stomach\/duodenum volume change was 28% less than for LDDMM (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0=\u00a00.0001). RMSD of DL\u2010predicted D(5cc) in small bowel was 39% less than for MMFF (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0=\u00a00.001); in stomach\/duodenum, RMSD of DL\u2010predicted D(5cc) was 18% less than for LDDMM (\n                      <jats:italic>P<\/jats:italic>\n                      \u00a0&lt;\u00a010\n                      <jats:sup>\u22123<\/jats:sup>\n                      ).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed deep network CT\u2010to\u2010CBCT deformable registration model shows improved segmentation accuracy compared to intensity\u2010based algorithms and achieves an order\u2010of\u2010magnitude reduction in processing time.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.14906","type":"journal-article","created":{"date-parts":[[2021,4,27]],"date-time":"2021-04-27T16:56:30Z","timestamp":1619542590000},"page":"3084-3095","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":35,"title":["Deep\u2010learning\u2010based image registration and automatic segmentation of organs\u2010at\u2010risk in cone\u2010beam CT scans from high\u2010dose radiation treatment of pancreatic cancer"],"prefix":"10.1002","volume":"48","author":[{"given":"Xu","family":"Han","sequence":"first","affiliation":[{"name":"Department of Computer Science University of North Carolina Chapel Hill NC 27599 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Hong","sequence":"additional","affiliation":[{"name":"Department of Medical Physics Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marsha","family":"Reyngold","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Crane","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John","family":"Cuaron","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carla","family":"Hajj","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Justin","family":"Mann","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melissa","family":"Zinovoy","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hastings","family":"Greer","sequence":"additional","affiliation":[{"name":"Department of Computer Science University of North Carolina Chapel Hill NC 27599 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ellen","family":"Yorke","sequence":"additional","affiliation":[{"name":"Department of Medical Physics Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gig","family":"Mageras","sequence":"additional","affiliation":[{"name":"Department of Medical Physics Memorial Sloan Kettering Cancer Center New York NY 10065 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marc","family":"Niethammer","sequence":"additional","affiliation":[{"name":"Department of Computer Science University of North Carolina Chapel Hill NC 27599 USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2021,5,14]]},"reference":[{"key":"e_1_2_9_2_1","unstructured":"American Cancer Society.Cancer Facts and Figures 2020.https:\/\/www.cancer.org\/research\/cancer\u2010facts\u2010statistics\/all\u2010cancer\u2010facts\u2010figures\/cancer\u2010facts\u2010figures\u20102020.html. Accessed December 4 2020."},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2015.12.003"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-019-1309-x"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00435"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1023\/B:VISI.0000043755.93987.aa"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/83.701170"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2001.990497"},{"key":"e_1_2_9_9_1","volume-title":"Numerical methods for image registration","author":"Modersitzki J","year":"2004"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2008.10.040"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1090\/qam\/1632326"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2008.10.050"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2017.07.008"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2897538"},{"key":"e_1_2_9_15_1","first-page":"2017","article-title":"Spatial transformer networks","author":"Jaderberg M","year":"2015","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_9_17_1","article-title":"PyTorch: An imperative style, high\u2010performance deep learning library","author":"Paszke A","year":"2019","journal-title":"arXiv preprint"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13566"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14386"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14906","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.14906","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14906","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,23]],"date-time":"2023-08-23T15:09:44Z","timestamp":1692803384000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.14906"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5,14]]},"references-count":18,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2021,6]]}},"alternative-id":["10.1002\/mp.14906"],"URL":"https:\/\/doi.org\/10.1002\/mp.14906","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5,14]]},"assertion":[{"value":"2020-12-10","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-04-08","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-05-14","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}