% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Schneider:300196,
author = {M. Schneider and S. Gutwein and D. Mönnich and C. Gani and
P. Fischer and C. F. Baumgartner and D. Thorwarth$^*$},
title = {{D}evelopment and comprehensive clinical validation of a
deep neural network for radiation dose modelling to enhance
magnetic resonance imaging guided radiotherapy.},
journal = {Physics $\&$ Imaging in Radiation Oncology},
volume = {33},
issn = {2405-6316},
address = {Amsterdam [u. a.]},
publisher = {Elsevier Science},
reportid = {DKFZ-2025-00673},
pages = {100723},
year = {2025},
abstract = {Online adaptive magnetic resonance imaging (MRI)-guided
radiotherapy requires fast dose calculation algorithms to
reduce intra-fraction motion uncertainties and improve
workflow efficiency. While Monte-Carlo simulations are
precise but computationally intensive, neural networks
promise fast and accurate dose modelling in strong magnetic
fields. This study aimed to train and evaluate a deep neural
network for dose modelling in MRI-guided radiotherapy using
a comprehensive clinical dataset.A dataset of 6595 clinical
irradiation segments from 125 1.5 T MRI-Linac radiotherapy
plans for various tumors sites was used. A 3D U-Net was
trained with 3961 segments using 3D imaging data and field
parameters as input, Root Mean Squared Error and a custom
loss function, with full Monte-Carlo simulations as ground
truth. For 2656 segments from 50 patients, gamma pass rates
(γ-PR) for 3 $mm/3\%,$ 2 $mm/2\%,$ and 1 $mm/1\%$ criteria
were calculated to assess dose modelling accuracy.
Performance was also tested in a standardized water phantom
to evaluate basic radiation physics properties.The neural
network accurately modeled dose distributions in both
patient and water phantom settings. Median (range) γ-PR of
$97.7\%$ $(87.5-100.0\%),$ $89.1\%$ $(69.7-99.4\%),$ and
$60.8\%$ $(38.5-82.1\%)$ were observed for treatment plans,
and $97.1\%$ $(55.5-100.0\%),$ $88.8\%$ $(38.8-99.7\%),$ and
$61.7\%$ $(17.9-94.4\%)$ for individual segments, across the
three criteria.High median γ-PR and accurate modelling in
both water phantom and clinical data demonstrate the high
potential of neural networks for dose modelling. However,
instances of lower γ-PR highlight the need for
comprehensive test data, improved robustness and future
built-in uncertainty estimation.},
keywords = {Artificial intelligence (Other) / Dose calculation (Other)
/ MRI-guided radiotherapy (Other) / Online adaptive
radiotherapy (Other)},
cin = {TU01},
ddc = {610},
cid = {I:(DE-He78)TU01-20160331},
pnm = {899 - ohne Topic (POF4-899)},
pid = {G:(DE-HGF)POF4-899},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:40093656},
pmc = {pmc:PMC11908596},
doi = {10.1016/j.phro.2025.100723},
url = {https://inrepo02.dkfz.de/record/300196},
}