{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T07:12:13Z","timestamp":1778137933864,"version":"3.51.4"},"reference-count":39,"publisher":"IOP Publishing","issue":"9","license":[{"start":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T00:00:00Z","timestamp":1778112000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T00:00:00Z","timestamp":1778112000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100004543","name":"China Scholarship Council","doi-asserted-by":"crossref","award":["202308440107"],"award-info":[{"award-number":["202308440107"]}],"id":[{"id":"10.13039\/501100004543","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["469106425"],"award-info":[{"award-number":["469106425"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2026,5,14]]},"abstract":"Abstract<\/jats:title>\n \n Objective.<\/jats:italic>\n Deep learning (DL) methods enable photon dose calculation under two main coordinate representations: Beam\u2019s Eye View (BEV) and patient coordinates. We evaluate dose calculation accuracy and speed under these coordinate paradigms and with representative DL models within a unified dataset and pipeline, and introduce two lightweight models for fast photon dose calculation.\n Approach.<\/jats:italic>\n Planning computed tomography (CT) scans and volumetric modulated arc therapy plans from 24 prostate cancer patients were used. Monte Carlo simulation generated 5940, 540, and 3053 segment doses for training (11 patients), validation (3), and testing (10), respectively. For BEV, we used a combination of convolutional neural network (CNN) and convolutional long short-term memory network (ConvLSTM) called CNN-ConvLSTM, a CNN-Mamba combination (CNN-Mamba), a transformer-based architecture (DoTA), and a cascaded 3D UNet (C3D). These were trained on CT and segment-projection BEV cuboids. For patient coordinates, the DeepDose individual segment dose prediction framework implemented with C3D (DeepDose-C3D) was trained on cropped CT volumes with four physical inputs. Segment and plan dose accuracy were assessed using local gamma passing rates\n \n \n \n <\/jats:tex-math>\n \n \n \n \u03b3<\/mml:mi>\n \n \n PR<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n (2%\/3\u2009mm and 1%\/3\u2009mm) and dose-volume histogram metrics. Dose calculation times (inference plus pre\/post-processing) were measured on three different graphics processing unit (GPUs).\n Results.<\/jats:italic>\n All five models achieved mean local\n \n \n \n <\/jats:tex-math>\n \n \n \n \u03b3<\/mml:mi>\n \n \n PR<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n values\n \n \n \n <\/jats:tex-math>\n \n \n \u2a7e<\/mml:mtext>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n 91.0% (2%\/3\u2009mm) for segment doses and\n \n \n \n <\/jats:tex-math>\n \n \n \u2a7e<\/mml:mtext>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n 99.0% (1%\/3\u2009mm) for plan doses. Mean per-segment dose calculation times were 79, 67, 298, 490, 356\u2009ms for CNN-ConvLSTM, CNN-Mamba, DoTA, C3D, and DeepDose-C3D, respectively. On the latest-generation GPU available, the corresponding per-plan (average 305 segments) dose calculation times were 5.5, 6.2, 33.6, 38.7, 35.4\u2009s.\n Significance.<\/jats:italic>\n Both BEV- and patient-coordinate DL methods achieved accurate photon plan dose calculation, with BEV-based approaches showing more robust segment performance. CNN-ConvLSTM and CNN-Mamba retain comparable accuracy at lower computational cost, enabling fast photon dose calculation.\n <\/jats:p>","DOI":"10.1088\/1361-6560\/ae6413","type":"journal-article","created":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T22:52:47Z","timestamp":1776984767000},"page":"095015","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Multi-model study of fast VMAT segment dose calculation with deep learning"],"prefix":"10.1088","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7502-0730","authenticated-orcid":true,"given":"Fan","family":"Xiao","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1451-223X","authenticated-orcid":true,"given":"Niklas","family":"Wahl","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Claus","family":"Belka","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Christopher","family":"Kurz","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"George","family":"Dedes","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1707-4068","authenticated-orcid":true,"given":"Guillaume","family":"Landry","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2026,5,7]]},"reference":[{"key":"pmbae6413bib1","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/abdbfe","type":"journal-article","article-title":"Deep dose plugin: towards real-time Monte Carlo dose calculation through a deep learning-based denoising algorithm","volume":"2","author":"Bai","year":"2021","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"pmbae6413bib2","doi-asserted-by":"publisher","first-page":"4555","DOI":"10.21105\/joss.04555","type":"journal-article","article-title":"Pymedphys: a community effort to develop an open, python-based standard library for medical physics applications","volume":"7","author":"Biggs","year":"2022","journal-title":"J. Open Source Softw."},{"key":"pmbae6413bib3","article-title":"Phase-space database for external beam radiotherapy. Summary report of a consultants\u2019 meeting","author":"Capote","year":"2006","type":"other"},{"key":"pmbae6413bib4","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17822","type":"journal-article","article-title":"Automatic, machine-agnostic, convolution-based beam and fluence modeling for Monte Carlo independent dose calculation","volume":"52","author":"Chen","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae6413bib5","doi-asserted-by":"publisher","first-page":"3172","DOI":"10.1002\/mp.16337","type":"journal-article","article-title":"Development and clinical application of a GPU-based Monte Carlo dose verification module and software for 1.5 T MR-LINAC","volume":"50","author":"Cheng","year":"2023","journal-title":"Med. Phys."},{"key":"pmbae6413bib6","doi-asserted-by":"publisher","first-page":"5039","DOI":"10.1002\/mp.17903","type":"journal-article","article-title":"A novel dose calculation system implemented in image domain","volume":"52","author":"Fan","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae6413bib7","doi-asserted-by":"publisher","first-page":"3728","DOI":"10.1002\/mp.13002","type":"journal-article","article-title":"First clinical implementation of real-time, real anatomy tracking and radiation beam control","volume":"45","author":"Green","year":"2018","journal-title":"Med. Phys."},{"key":"pmbae6413bib8","article-title":"Mamba: linear-time sequence modeling with selective state spaces","author":"Gu","year":"2024","type":"conference-proceedings"},{"key":"pmbae6413bib9","doi-asserted-by":"publisher","first-page":"754","DOI":"10.1118\/1.3539725","type":"journal-article","article-title":"GPUMCD: a new GPU-oriented Monte Carlo dose calculation platform","volume":"38","author":"Hissoiny","year":"2011","journal-title":"Med. Phys."},{"key":"pmbae6413bib10","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20180270","type":"journal-article","article-title":"Automation in intensity modulated radiotherapy treatment planning-a review of recent innovations","volume":"91","author":"Hussein","year":"2018","journal-title":"Br. J. Radiol."},{"key":"pmbae6413bib11","doi-asserted-by":"publisher","first-page":"787","DOI":"10.1016\/j.ijrobp.2025.04.019","type":"journal-article","article-title":"Critical review: real-time dose-guided radiation therapy","volume":"122","author":"Keall","year":"2025","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"pmbae6413bib12","doi-asserted-by":"publisher","first-page":"7485","DOI":"10.1088\/0031-9155\/60\/19\/7485","type":"journal-article","article-title":"A new methodology for inter-and intrafraction plan adaptation for the MR-linac","volume":"60","author":"Kontaxis","year":"2015","journal-title":"Phys. Med. Biol."},{"key":"pmbae6413bib13","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab7630","type":"journal-article","article-title":"DeepDose: Towards a fast dose calculation engine for radiation therapy using deep learning","volume":"65","author":"Kontaxis","year":"2020","journal-title":"Phys. Med. Biol."},{"key":"pmbae6413bib14","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2025.100806","type":"journal-article","article-title":"A proof-of-concept study of direct magnetic resonance imaging-based proton dose calculation for brain tumors via neural networks with Monte Carlo-comparable accuracy","volume":"35","author":"Li","year":"2025","journal-title":"Phys. Imaging Radiat. Oncol."},{"key":"pmbae6413bib15","doi-asserted-by":"publisher","first-page":"231","DOI":"10.1007\/s40846-024-00863-x","type":"journal-article","article-title":"Role of artificial intelligence in medical image analysis: A review of current trends and future directions","volume":"44","author":"Li","year":"2024","journal-title":"J. Med. Biol. Eng."},{"key":"pmbae6413bib16","doi-asserted-by":"publisher","first-page":"6174","DOI":"10.1002\/mp.15165","type":"journal-article","article-title":"Extension and validation of a GPU-Monte Carlo dose engine gDPM for 1.5 T MR-LINAC online independent dose verification","volume":"48","author":"Li","year":"2021","journal-title":"Med. Phys."},{"key":"pmbae6413bib17","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1186\/s13014-024-02534-2","type":"journal-article","article-title":"A deep learning-based dose calculation method for volumetric modulated arc therapy","volume":"19","author":"Liang","year":"2024","journal-title":"Radiat. Oncol."},{"key":"pmbae6413bib18","doi-asserted-by":"publisher","first-page":"5574","DOI":"10.1002\/mp.15034","type":"journal-article","article-title":"A cascade 3D U-Net for dose prediction in radiotherapy","volume":"48","author":"Liu","year":"2021","journal-title":"Med. Phys."},{"key":"pmbae6413bib19","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17899","type":"journal-article","article-title":"A GPU-accelerated Monte Carlo dose engine for external beam radiotherapy","volume":"52","author":"Liu","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae6413bib20","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2023.109970","type":"journal-article","article-title":"Real-time motion management in MRI-guided radiotherapy: current status and AI-enabled prospects","volume":"190","author":"Lombardo","year":"2024","journal-title":"Radiother. Oncol."},{"key":"pmbae6413bib21","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1016\/j.ejmp.2019.06.007","type":"journal-article","article-title":"Optimization of phase space files from clinical linear accelerators","volume":"64","author":"Martins","year":"2019","journal-title":"Phys. Med."},{"key":"pmbae6413bib22","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2025.109777","type":"journal-article","article-title":"Real-time adaptive proton therapy: an AI-based spatio-temporal mono-energetic dose calculation model (CC-LSTM)","volume":"188","author":"Neishabouri","year":"2025","journal-title":"Comput. Biol. Med."},{"key":"pmbae6413bib23","article-title":"Cupy: a numpy-compatible library for nvidia gpu calculations","author":"Okuta","year":"2017","type":"conference-proceedings"},{"key":"pmbae6413bib24","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/j.ejmp.2018.06.003","type":"journal-article","article-title":"Monte Carlo dose verification of VMAT treatment plans using Elekta Agility 160-leaf MLC","volume":"51","author":"Onizuka","year":"2018","journal-title":"Phys. Med."},{"key":"pmbae6413bib25","doi-asserted-by":"publisher","first-page":"310","DOI":"10.1118\/1.2818738","type":"journal-article","article-title":"Volumetric modulated arc therapy: IMRT in a single gantry arc","volume":"35","author":"Otto","year":"2008","journal-title":"Med. Phys."},{"key":"pmbae6413bib26","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1016\/j.radonc.2009.05.010","type":"journal-article","article-title":"AAA and PBC calculation accuracy in the surface build-up region in tangential beam treatments. Phantom and breast case study with the Monte Carlo code PENELOPE","volume":"93","author":"Panettieri","year":"2009","journal-title":"Radiother. Oncol."},{"key":"pmbae6413bib27","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13715","type":"journal-article","article-title":"Monte Carlo modeling of the Elekta Versa HD and patient dose calculation with EGSnrc\/BEAMnrc","volume":"23","author":"Paschal","year":"2022","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"pmbae6413bib28","doi-asserted-by":"publisher","first-page":"3159","DOI":"10.1002\/mp.16231","type":"journal-article","article-title":"Sub-second photon dose prediction via transformer neural networks","volume":"50","author":"Pastor-Serrano","year":"2023","journal-title":"Med. Phys."},{"key":"pmbae6413bib29","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1007\/s00066-024-02277-9","type":"journal-article","article-title":"Artificial intelligence for treatment delivery: image-guided radiotherapy","volume":"201","author":"Rabe","year":"2025","journal-title":"Strahlenther. Onkol."},{"key":"pmbae6413bib30","doi-asserted-by":"publisher","first-page":"479","DOI":"10.1118\/1.4903502","type":"journal-article","article-title":"Latent uncertainties of the precalculated track Monte Carlo method","volume":"42","author":"Renaud","year":"2015","journal-title":"Med. Phys."},{"key":"pmbae6413bib31","doi-asserted-by":"publisher","first-page":"9329","DOI":"10.1088\/0031-9155\/60\/24\/9329","type":"journal-article","article-title":"Monte Carlo study on the sensitivity of prompt gamma imaging to proton range variations due to interfractional changes in prostate cancer patients","volume":"60","author":"Schmid","year":"2015","journal-title":"Phys. Med. Biol."},{"key":"pmbae6413bib32","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2025.100723","type":"journal-article","article-title":"Development and comprehensive clinical validation of a deep neural network for radiation dose modelling to enhance magnetic resonance imaging guided radiotherapy","volume":"33","author":"Schneider","year":"2025","journal-title":"Phys. Imaging Radiat. Oncol."},{"key":"pmbae6413bib33","first-page":"p 28","type":"conference-proceedings","article-title":"Convolutional LSTM network: A machine learning approach for precipitation nowcasting","author":"Shi","year":"2015"},{"key":"pmbae6413bib34","doi-asserted-by":"publisher","first-page":"2429","DOI":"10.1002\/mp.16077","type":"journal-article","article-title":"Cross-engine transformation-based fast dose calculation for MRI-Linac online treatment planning","volume":"50","author":"Song","year":"2023","journal-title":"Med. Phys."},{"key":"pmbae6413bib35","doi-asserted-by":"publisher","DOI":"10.1016\/j.phro.2024.100575","type":"journal-article","article-title":"A deep learning based dynamic arc radiotherapy photon dose engine trained on Monte Carlo dose distributions","volume":"30","author":"Witte","year":"2024","journal-title":"Phys. Imaging Radiat. Oncol."},{"key":"pmbae6413bib36","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac7376","type":"journal-article","article-title":"TransDose: a transformer-based UNet model for fast and accurate dose calculation for MR-LINACs","volume":"67","author":"Xiao","year":"2022","journal-title":"Phys. Med. Biol."},{"key":"pmbae6413bib37","doi-asserted-by":"publisher","DOI":"10.1002\/mp.70106","type":"journal-article","article-title":"Deep learning-based synthetic-ct-free photon dose calculation in mr-guided radiotherapy: a proof-of-concept study","volume":"52","author":"Xiao","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae6413bib38","first-page":"pp 1","type":"conference-proceedings","article-title":"Reluplex made more practical: Leaky relu","author":"Xu","year":"2020"},{"key":"pmbae6413bib39","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ad829e","type":"journal-article","article-title":"Quality assurance for online adaptive radiotherapy: a secondary dose verification model with geometry-encoded u-net","volume":"5","author":"Yan","year":"2024","journal-title":"Mach. Learn.: Sci. Technol."}],"container-title":["Physics in Medicine & Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,7]],"date-time":"2026-05-07T06:32:47Z","timestamp":1778135567000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae6413"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,7]]},"references-count":39,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,5,7]]},"published-print":{"date-parts":[[2026,5,14]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/ae6413","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,7]]},"assertion":[{"value":"Multi-model study of fast VMAT segment dose calculation with deep learning","name":"article_title","label":"Article Title"},{"value":"Physics in Medicine & Biology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 The Author(s). Published on behalf of Institute of Physics and Engineering in Medicine by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-12-03","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-04-23","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-05-07","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}