%0 Journal Article
%A Huang, L.
%A Thummerer, A.
%A Papadopoulou, C. I.
%A Corradini, S.
%A Belka, C.
%A Riboldi, M.
%A Kurz, C.
%A Landry, G.
%T Validation of patient-specific deep learning markerless lung tumor tracking aided by 4DCBCT.
%J Physics in medicine and biology
%V 70
%N 6
%@ 0031-9155
%C Bristol
%I IOP Publ.
%M DKFZ-2025-00529
%P 065001
%D 2025
%X Objective. 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. 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
%K Lung Neoplasms: diagnostic imaging
%K Humans
%K Deep Learning
%K Cone-Beam Computed Tomography: methods
%K Four-Dimensional Computed Tomography: methods
%K Image Processing, Computer-Assisted: methods
%K 4DCBCT (Other)
%K AI (Other)
%K lung tumor targeting (Other)
%K markerless tracking (Other)
%K patient-specific (Other)
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:39978071
%R 10.1088/1361-6560/adb89c
%U https://inrepo02.dkfz.de/record/299588