{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T17:46:23Z","timestamp":1784569583139,"version":"3.55.0"},"reference-count":30,"publisher":"IOP Publishing","issue":"6","license":[{"start":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T00:00:00Z","timestamp":1741564800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T00:00:00Z","timestamp":1741564800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2025,3,16]]},"abstract":"Abstract<\/jats:title>\n \n Objective<\/jats:italic>. Tracking tumors with multi-leaf collimators and x-ray imaging can be a cost-effective motion management method to reduce internal target volume margins for lung cancer patients, sparing normal tissues while ensuring target coverage. To realize that, accurate tumor localization on x-ray images is essential. We aimed to develop a systematic method for automatically generating tumor segmentation ground truth (GT) on cone-beam computed tomography (CBCT) projections and use it to help refine and validate our patient-specific AI-based tumor localization model. Approach<\/jats:italic>. To obtain the tumor segmentation GT on CBCT projections, we propose a 4DCBCT-aided GT generation pipeline consisting of three steps: breathing phase extraction and 10-phase 4DCBCT reconstruction, manual segmentation on phase 50% followed by deformable contour propagation to other phases, and forward projection of the 3D segmentation to the CBCT projection of the corresponding phase. We then used the CBCT projections from one fraction in the angular range of [\n \n \n \n \u2212<\/mml:mo>\n \n 10<\/mml:mn>\n \n \u2218<\/mml:mo>\n <\/mml:mrow>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>, 10\u2218<\/jats:sup>] and [80\u2218<\/jats:sup>, 100\u2218<\/jats:sup>] to refine a Retina U-Net baseline model, which was pretrained on 1140231 digitally reconstructed radiographs generated from a public lung dataset for automatic tumor delineation on projections, and used later-fraction CBCT projections in the same angular range for testing. Six LMU University Hospital patient CBCT projection sets were reserved for validation and 11 for testing. Tracking accuracy was evaluated as the center-of-mass (COM) error and the Dice similarity coefficient (DSC) between the predicted and ground-truth segmentations. Main results<\/jats:italic>. Over the 11 testing patients, each with around 40 CBCT projections tested, the patient refined models had a mean COM error of 2.3 \u00b1 0.9 mm\/4.2 \u00b1 1.7 mm and a mean DSC of 0.83 \u00b1 0.06\/0.72 \u00b1 0.13 for angles within [\n \n \n \n \u2212<\/mml:mo>\n \n 10<\/mml:mn>\n \n \u2218<\/mml:mo>\n <\/mml:mrow>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>, 10\u2218<\/jats:sup>] \/ [80\u2218<\/jats:sup>, 100\u2218<\/jats:sup>]. The mean inference time was 68 ms\/frame. The patient-specific training segmentation loss was found to be correlated to the segmentation performance at [\n \n \n \n \u2212<\/mml:mo>\n \n 10<\/mml:mn>\n \n \u2218<\/mml:mo>\n <\/mml:mrow>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>, 10\u2218<\/jats:sup>]. Significance<\/jats:italic>. Our proposed approach allows patient-specific real-time markerless lung tumor tracking, which could be validated thanks to the novel 4DCBCT-aided GT generation approach.<\/jats:p>","DOI":"10.1088\/1361-6560\/adb89c","type":"journal-article","created":{"date-parts":[[2025,2,20]],"date-time":"2025-02-20T22:55:05Z","timestamp":1740092105000},"page":"065001","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Validation of patient-specific deep learning markerless lung tumor tracking aided by 4DCBCT"],"prefix":"10.1088","volume":"70","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-2437-9843","authenticated-orcid":true,"given":"L","family":"Huang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1874-5030","authenticated-orcid":true,"given":"A","family":"Thummerer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C I","family":"Papadopoulou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S","family":"Corradini","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C","family":"Belka","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"M","family":"Riboldi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"C","family":"Kurz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1707-4068","authenticated-orcid":true,"given":"G","family":"Landry","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2025,3,10]]},"reference":[{"key":"pmbadb89cbib1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102998","article-title":"Volumetric tumor tracking from a single cone-beam x-ray projection image enabled by deep learning","volume":"91","author":"Dai","year":"2024","journal-title":"Med. Image Anal."},{"key":"pmbadb89cbib2","doi-asserted-by":"publisher","first-page":"213","DOI":"10.1016\/j.radonc.2020.10.004","article-title":"RealDRR \u2013 rendering of realistic digitally reconstructed radiographs using locally trained image-to-image translation","volume":"153","author":"Dhont","year":"2020","journal-title":"Radiother. Oncol."},{"key":"pmbadb89cbib3","doi-asserted-by":"publisher","first-page":"5953","DOI":"10.1002\/mp.14445","article-title":"DICOM re-encoding of volumetrically annotated Lung Imaging Database Consortium (LIDC) nodules","volume":"47","author":"Fedorov","year":"2020","journal-title":"Med. Phys."},{"key":"pmbadb89cbib4","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1002\/acm2.12978","article-title":"Evaluation of radixact motion synchrony for 3D respiratory motion: modeling accuracy and dosimetric fidelity","volume":"21","author":"Ferris","year":"2020","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"pmbadb89cbib5","doi-asserted-by":"publisher","first-page":"6881","DOI":"10.1002\/mp.16470","article-title":"Deep learning-based markerless lung tumor tracking in stereotactic radiotherapy using Siamese networks","volume":"50","author":"Grama","year":"2023","journal-title":"Med. Phys."},{"key":"pmbadb89cbib6","doi-asserted-by":"publisher","first-page":"1957","DOI":"10.1002\/mp.16705","article-title":"Simultaneous object detection and segmentation for patient-specific markerless lung tumor tracking in simulated radiographs with deep learning","volume":"51","author":"Huang","year":"2024","journal-title":"Med. Phys."},{"key":"pmbadb89cbib7","first-page":"pp 171","article-title":"Retina U-Net: embarrassingly simple exploitation of segmentation supervision for medical object detection","author":"Jaeger","year":"2020"},{"key":"pmbadb89cbib8","doi-asserted-by":"publisher","DOI":"10.1016\/j.patter.2023.100804","article-title":"Leakage and the reproducibility crisis in machine-learning-based science","volume":"4","author":"Kapoor","year":"2023","journal-title":"Patterns"},{"key":"pmbadb89cbib9","doi-asserted-by":"publisher","first-page":"5744","DOI":"10.1088\/1361-6560\/aa6ed7","article-title":"Quantifying the accuracy and precision of a novel real-time 6 degree-of-freedom kilovoltage intrafraction monitoring (KIM) target tracking system","volume":"62","author":"Kim","year":"2017","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib10","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/acb4d7","article-title":"Deep learning based direct segmentation assisted by deformable image registration for cone-beam CT based auto-segmentation for adaptive radiotherapy","volume":"68","author":"Liang","year":"2023","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib11","doi-asserted-by":"publisher","first-page":"6856","DOI":"10.1088\/0031-9155\/61\/18\/6856","article-title":"Motion-aware temporal regularization for improved 4D cone-beam computed tomography","volume":"61","author":"Mory","year":"2016","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib12","first-page":"pp 407","article-title":"Deformable image registration of sliding organs using anisotropic diffusive regularization","author":"Pace","year":"2011"},{"key":"pmbadb89cbib13","doi-asserted-by":"publisher","first-page":"419","DOI":"10.1016\/j.radonc.2015.05.004","article-title":"A comparison of two clinical correlation models used for real-time tumor tracking of semi-periodic motion: a focus on geometrical accuracy in lung and liver cancer patients","volume":"115","author":"Poels","year":"2015","journal-title":"Radiother. Oncol."},{"key":"pmbadb89cbib14","doi-asserted-by":"publisher","first-page":"1587","DOI":"10.1016\/j.ijrobp.2008.07.037","article-title":"A method to estimate mean position, motion magnitude, motion correlation and trajectory of a tumor from cone-beam CT projections for image-guided radiotherapy","volume":"72","author":"Poulsen","year":"2008","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"pmbadb89cbib15","doi-asserted-by":"publisher","DOI":"10.1088\/1742-6596\/489\/1\/012079","article-title":"The Reconstruction Toolkit (RTK), an open-source cone-beam CT reconstruction toolkit based on the Insight Toolkit (ITK)","volume":"489","author":"Rit","year":"2014","journal-title":"J. Phys.: Conf. Ser."},{"key":"pmbadb89cbib16","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab79c5","article-title":"A machine learning-based real-time tumor tracking system for fluoroscopic gating of lung radiotherapy","volume":"65","author":"Sakata","year":"2020","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib17","doi-asserted-by":"publisher","first-page":"1107","DOI":"10.3109\/0284186X.2014.927585","article-title":"Clinical use of iterative 4D-cone beam computed tomography reconstructions to investigate respiratory tumor motion in lung cancer patients","volume":"53","author":"Schmidt","year":"2014","journal-title":"Acta Oncol."},{"key":"pmbadb89cbib18","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac16e9","article-title":"Validation of proton dose calculation on scatter corrected 4D cone beam computed tomography using a porcine lung phantom","volume":"66","author":"Schmitz","year":"2021","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib19","doi-asserted-by":"publisher","first-page":"4981","DOI":"10.1002\/mp.16335","article-title":"Scatter correction of 4D cone beam computed tomography to detect dosimetric effects due to anatomical changes in proton therapy for lung cancer","volume":"50","author":"Schmitz","year":"2023","journal-title":"Med. Phys."},{"key":"pmbadb89cbib20","article-title":"Plastimatch: an open source software suite for radiotherapy image processing","author":"Sharp","year":"2010"},{"key":"pmbadb89cbib21","doi-asserted-by":"publisher","first-page":"3065","DOI":"10.1088\/1361-6560\/aa6393","article-title":"A Bayesian approach for three-dimensional markerless tumor tracking using kV imaging during lung radiotherapy","volume":"62","author":"Shieh","year":"2017","journal-title":"Phys. Med. Biol."},{"key":"pmbadb89cbib22","doi-asserted-by":"publisher","first-page":"1176","DOI":"10.1118\/1.1869074","article-title":"Respiratory correlated cone beam CT","volume":"32","author":"Sonke","year":"2005","journal-title":"Med. Phys."},{"key":"pmbadb89cbib23","doi-asserted-by":"publisher","first-page":"932","DOI":"10.1016\/j.ijrobp.2018.04.056","article-title":"4-dimensional cone beam computed tomography\u2013measured target motion underrepresents actual motion","volume":"102","author":"Steiner","year":"2018","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"pmbadb89cbib24","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20190420","article-title":"Real-time markerless tumour tracking with patient-specific deep learning using a personalised data generation strategy: proof of concept by phantom study","volume":"93","author":"Takahashi","year":"2020","journal-title":"Br. J. Radiol."},{"key":"pmbadb89cbib25","doi-asserted-by":"publisher","first-page":"480","DOI":"10.1002\/mp.16095","article-title":"Explainability and controllability of patient-specific deep learning with attention-based augmentation for markerless image-guided radiotherapy","volume":"50","author":"Terunuma","year":"2023","journal-title":"Med. Phys."},{"key":"pmbadb89cbib26","doi-asserted-by":"publisher","first-page":"2540","DOI":"10.1118\/1.4917480","article-title":"Real-time markerless lung tumor tracking in fluoroscopic video: Handling overlapping of projected structures","volume":"42","author":"Teske","year":"2015","journal-title":"Med. Phys."},{"key":"pmbadb89cbib27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-00937-3_12","article-title":"DeepDRR \u2013 a catalyst for machine learning in fluoroscopy-guided procedures","author":"Unberath","year":"2018"},{"key":"pmbadb89cbib28","doi-asserted-by":"publisher","first-page":"37026","DOI":"10.1109\/ACCESS.2019.2899385","article-title":"Convolutional neural network (CNN) based three dimensional tumor localization using single x-ray projection","volume":"7","author":"Wei","year":"2019","journal-title":"IEEE Access"},{"key":"pmbadb89cbib29","doi-asserted-by":"publisher","first-page":"432","DOI":"10.1016\/j.ijrobp.2019.05.071","article-title":"Markerless pancreatic tumor target localization enabled by deep learning","volume":"105","author":"Zhao","year":"2019","journal-title":"Int. J. Radiat. Oncol. Biol. Phys."},{"key":"pmbadb89cbib30","doi-asserted-by":"publisher","first-page":"1382","DOI":"10.1002\/mp.15456","article-title":"Development of a deep learning-based patient-specific target contour prediction model for markerless tumor positioning","volume":"49","author":"Zhou","year":"2022","journal-title":"Med. Phys."}],"container-title":["Physics in Medicine & Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T07:03:11Z","timestamp":1741590191000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/adb89c"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,10]]},"references-count":30,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2025,3,10]]},"published-print":{"date-parts":[[2025,3,16]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/adb89c","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,10]]},"assertion":[{"value":"Validation of patient-specific deep learning markerless lung tumor tracking aided by 4DCBCT","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 2025 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":"2024-08-28","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-02-20","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-03-10","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}