{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T13:45:29Z","timestamp":1792071929058,"version":"4.4.0"},"reference-count":41,"publisher":"Wiley","issue":"9","license":[{"start":{"date-parts":[[2026,8,30]],"date-time":"2026-08-30T00:00:00Z","timestamp":1788048000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,8,30]],"date-time":"2026-08-30T00:00:00Z","timestamp":1788048000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001691","name":"Japan Society for the Promotion of Science","doi-asserted-by":"publisher","award":["23K24282"],"award-info":[{"award-number":["23K24282"]}],"id":[{"id":"10.13039\/501100001691","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":[[2026,9]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Pancreatic cancer Volumetric Modulated Arc Therapy (VMAT) planning presents a significant dosimetric challenge due to the high\u2010dose gradients required to spare adjacent, radiosensitive organs at risk (OARs) like the stomach and duodenum. This anatomical complexity has limited the scope of automated planning for this site. Broadly, while deep learning (DL) has been introduced to streamline treatment planning, most existing models only predict intermediate outputs, such as dose distributions or fluence maps, which still necessitate a subsequent, computationally expensive inverse optimization step on a treatment planning system.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To address these challenges, we aim to develop an optimization\u2010free fully automated VMAT planning framework for pancreatic cancer. As a key component of this system, this study introduces a DL model designed to directly generate machine parameters from dose distributions and anatomical contours.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      A total of 200 vmat plans for pancreatic cancer (prescription: 42\u00a0Gy in 15 fractions) were retrospectively collected. The dataset was randomly split into training (\n                      <jats:italic>n\u00a0<\/jats:italic>\n                      =\u00a0170), validation (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a010), and testing (\n                      <jats:italic>n<\/jats:italic>\n                      \u00a0=\u00a020) sets. The proposed Multi\u2010modal, Attention &amp; Transformer\u2010Enhanced U\u2010Net (MATE\u2010UNet) utilizes beam's\u2010eye\u2010view (BEV) projections of the reference dose distribution and anatomical contours to directly predict machine\u2010executable multi\u2010leaf collimator (MLC) apertures and Monitor Units (MUs). The proposed model was benchmarked against baseline U\u2010Nets (using contour\u2010only, dose\u2010only, and combined inputs) on the testing set. Model accuracy was assessed using the Dice Similarity Coefficient (DSC) for MLC and Mean Absolute Error (MAE) for MU, while plan quality was evaluated using clinical dose\u2010volume histogram (DVH) metrics and Conformity Index (CI), Homogeneity Index (HI), and Gradient Index (GI).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      MATE\u2010UNet achieved a clinical acceptance rate of 100% (20\/20), compared to 70% for the best\u2010performing baseline model. In terms of prediction accuracy, the proposed model achieved a DSC of 0.9553\u00a0\u00b1\u00a00.0042 for MLC apertures and an MAE of 1.960\u00a0\u00b1\u00a00.396 for MU. Dosimetric evaluation demonstrated that, relative to the reference plans, MATE\u2010UNet maintained comparable target coverage, CI (0.773), and HI (0.100), while achieving a significantly improved GI (3.732,\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.05). Furthermore, MATE\u2010UNet significantly reduced the V\n                      <jats:sub>39Gy<\/jats:sub>\n                      of the stomach and duodenum, as well as the global maximum dose, compared with the baseline models (\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0&lt;\u00a00.05).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>This study demonstrates the feasibility of MATE\u2010UNet for the direct prediction of VMAT machine parameters without iterative optimization. By leveraging multi\u2010modal BEV inputs and a Transformer\u2010enhanced architecture, the proposed framework represents a valuable step toward bridging the gap between dose distributions and clinically usable treatment plans. Although an additional normalization step is required to determine the absolute MUs, MATE\u2010UNet has the potential to serve as a downstream component of a future fully automated anatomy\u2010to\u2010plan pipeline.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70656","type":"journal-article","created":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T05:12:31Z","timestamp":1788153151000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Multi\u2010modal direct LINAC parameter prediction for pancreatic VMAT: An optimization\u2010free approach"],"prefix":"10.1002","volume":"53","author":[{"given":"Zixu","family":"Guan","sequence":"first","affiliation":[{"name":"Department of Advanced Medical Physics Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yukine","family":"Shimizu","sequence":"additional","affiliation":[{"name":"Department of Advanced Medical Physics Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takahiro","family":"Iwai","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Image\u2010Applied Therapy Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michio","family":"Yoshimura","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Image\u2010Applied Therapy Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takashi","family":"Mizowaki","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology and Image\u2010Applied Therapy Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mitsuhiro","family":"Nakamura","sequence":"additional","affiliation":[{"name":"Department of Advanced Medical Physics Graduate School of Medicine Kyoto University  Kyoto Japan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,8,30]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140\u20106736(25)00261\u20102"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj\u20102022\u2010073995"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.6004\/jnccn.2021.0017"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2019.06.016"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00066\u2010025\u201002433\u20109"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.2818738"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20201289"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2011.11.012"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2012.06.009"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.prro.2016.04.005"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13040667"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2024.110609"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010017\u20100808\u2010x"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13271"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/aa71f8"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17673"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmrs.420"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac80e5"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ad07f6"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16545"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2025.100724"},{"key":"e_1_2_8_23_1","doi-asserted-by":"crossref","unstructured":"RonnebergerO FischerP BroxT.U\u2010Net: convolutional networks for biomedical image segmentation.Springer International Publishing;2015:234\u2010241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_8_24_1","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek\u00d6 AbdulkadirA LienkampSS BroxT RonnebergerO.3D U\u2010Net: learning dense volumetric segmentation from sparse annotation.Springer;2016:424\u2010432.","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2018.2883958"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.3389\/fradi.2021.704888"},{"key":"e_1_2_8_27_1","doi-asserted-by":"crossref","unstructured":"HatamizadehA TangY NathV et\u00a0al.UNETR: transformers for 3D medical image segmentation.2022:1748\u20101758.","DOI":"10.1109\/WACV51458.2022.00181"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103280"},{"key":"e_1_2_8_29_1","unstructured":"MaasAL HannunAY NgAY.Rectifier nonlinearities improve neural network acoustic models.In: Proceedings of the 30th International Conference on Machine Learning;2013; Atlanta GA USA."},{"key":"e_1_2_8_30_1","unstructured":"OktayO SchlemperJ Le FolgocL et\u00a0al.Attention U\u2010Net: learning where to look for the pancreas.2018:arXiv:1804.03999 Accessed April 01 2018. doi:10.48550\/arXiv.1804.03999"},{"key":"e_1_2_8_31_1","unstructured":"VaswaniA ShazeerN ParmarN et\u00a0al.Attention is all you need.2017"},{"key":"e_1_2_8_32_1","unstructured":"HuJ ShenL AlbanieS SunG WuE Squeeze\u2010and\u2010excitation networks.2017:arXiv:1709.01507 Accessed September 01 2017. doi:10.48550\/arXiv.1709.01507"},{"key":"e_1_2_8_33_1","unstructured":"HendrycksD GimpelK Gaussian error linear units (gelus). arXiv preprint arXiv:160608415.2016"},{"key":"e_1_2_8_34_1","doi-asserted-by":"crossref","unstructured":"KendallA GalY CipollaR IEEE.Multi\u2010task learning using uncertainty to weigh losses for scene geometry and semantics.2018:7482\u20107491.","DOI":"10.1109\/CVPR.2018.00781"},{"key":"e_1_2_8_35_1","doi-asserted-by":"crossref","unstructured":"SalehiS ErdogmusD GholipourA Tversky loss function for image segmentation using 3D fully convolutional deep networks.2017:379\u2010387.","DOI":"10.1007\/978-3-319-67389-9_44"},{"key":"e_1_2_8_36_1","unstructured":"LoshchilovI HutterF Decoupled weight decay regularization. arXiv preprint arXiv:171105101.2017"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.3171\/jns.2000.93.supplement_3.0219"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1093\/jicru\/ndq002"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.3171\/sup.2006.105.7.194"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2022.05.008"},{"key":"e_1_2_8_41_1","doi-asserted-by":"crossref","unstructured":"\u00c7i\u00e7ek\u00d6 AbdulkadirA LienkampSS BroxT RonnebergerO 3D U\u2010Net: Learning dense volumetric segmentation from sparse annotation.2016:arXiv:1606.06650 Accessed June 01 2016. doi:10.48550\/arXiv.1606.06650","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.70291"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70656","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70656","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70656","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T12:55:40Z","timestamp":1792068940000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70656"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,30]]},"references-count":41,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2026,9]]}},"alternative-id":["10.1002\/mp.70656"],"URL":"https:\/\/doi.org\/10.1002\/mp.70656","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,30]]},"assertion":[{"value":"2026-01-29","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-10","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70656"}}