Background and purpose Adaptive radiotherapy based on cone-beam computed tomography (CBCT) requires high CT number accuracy to ensure accurate dose calculations. Recently, deep learning has been proposed for fast CBCT artefact corrections on single anatomical sites. This study investigated the feasibility of applying a single convolutional network to facilitate dose calculation based on CBCT for head-and-neck, lung and breast cancer patients. Materials and Methods Ninety-nine patients diagnosed with head-and-neck, lung or breast cancer undergoing radiotherapy with CBCT-based position verification were included in this study. The CBCTs were registered to planning CT according to clinical procedures. Three cycle-consistent generative adversarial networks (cycle-GANs) were trained in an unpaired manner on 15 patients per anatomical site generating synthetic-CTs (sCTs). Another network was trained with all the anatomical sites together. Performances of all four networks were compared and evaluated for image similarity against rescan CT (rCT). Clinical plans were recalculated on rCT and sCT and analysed through voxel-based dose differences and-analysis. Results A sCT was generated in 10 s. Image similarity was comparable between models trained on different anatomical sites and a single model for all sites. Mean dose differences <0.5% were obtained in high-dose regions. Mean gamma (3%, 3 mm) pass-rates were achieved for all sites. Conclusion Cycle-GAN reduced CBCT artefacts and increased similarity to CT, enabling sCT-based dose calculations. A single network achieved CBCT-based dose calculation generating synthetic CT for head-and-neck, lung, and breast cancer patients with similar performance to a network specifically trained for each anatomical site.
The UMC Utrecht MRI/linac (MRL) design provides image guidance with high soft-tissue contrast, directly during radiotherapy (RT). Breast cancer patients are a potential group to benefit from better guidance in the MRL. However, due to the electron return effect, the skin dose can be increased in presence of a magnetic field. Since large skin areas are generally involved in breast RT, the purpose of this study is to investigate the effects on the skin dose, for whole-breast irradiation (WBI) and accelerated partial-breast irradiation (APBI). In ten patients with early-stage breast cancer, targets and organs at risk (OARs) were delineated on postoperative CT scans co-registered with MRI. The OARs included the skin, comprising the first 5 mm of ipsilateral-breast tissue, plus extensions. Three intensity-modulated RT techniques were considered (2× WBI, 1× APBI). Individual beam geometries were used for all patients. Specially developed MRL treatment-planning software was used. Acceptable plans were generated for 0 T, 0.35 T and 1.5 T, using a class solution. The skin dose was augmented in WBI in the presence of a magnetic field, which is a potential drawback, whereas in APBI the induced effects were negligible. This opens possibilities for developing MR-guided partial-breast treatments in the MRL.
In early-stage breast-cancer patients, accelerated partial-breast irradiation techniques (APBI) and hypofractionation are increasingly implemented after breast-conserving surgery (BCS). For a safe and effective radiation therapy (RT), the influence of intra-fraction motion during dose delivery becomes more important as associated fraction durations increase and targets become smaller. Current image-guidance techniques are insufficient to characterize local target movement in high temporal and spatial resolution for extended durations. Magnetic resonance imaging (MRI) can provide high soft-tissue contrast, allow fast imaging, and acquire images during longer periods. The goal of this study was to quantify intra-fraction motion using MRI scans from 21 breast-cancer patients, before and after BCS, in supine RT position, on two time scales. High-temporal 2-dimensional (2D) MRI scans (cine-MRI), acquired every 0.3 s during 2 min, and three 3D MRI scans, acquired over 20 min, were performed. The tumor (bed) and whole breast were delineated on 3D scans and delineations were transferred to the cine-MRI series. Consecutive scans were rigidly registered and delineations were transformed accordingly. Motion in sub-second time-scale (derived from cine-MRI) was generally regular and limited to a median of 2 mm. Infrequently, large deviations were observed, induced by deep inspiration, but these were temporary. Movement on multi-minute scale (derived from 3D MRI) varied more, although medians were restricted to 2.2 mm or lower. Large whole-body displacements (up to 14 mm over 19 min) were sparsely observed. The impact of motion on standard RT techniques is likely small. However, in novel hypofractionated APBI techniques, whole-body shifts may affect adequate RT delivery, given the increasing fraction durations and smaller targets. Motion management may thus be required. For this, on-line MRI guidance could be provided by a hybrid MRI/RT modality, such as the University Medical Center Utrecht MRI linear accelerator.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.