Fast and accurate dose predictions for novel radiotherapy treatments in heterogeneous phantoms using conditional 3D‐UNet generative adversarial networks
Medical Physics2022Vol. 49(5), pp. 3389–3404
Citations Over TimeTop 10% of 2022 papers
Florian Mentzel, K. Kroeninger, Michael Lerch, O. Nackenhorst, Jason Paino, Anatoly Rosenfeld, Ayu Saraswati, Ah Chung Tsoi, J. Weingarten, Markus Hagenbuchner, Susanna Guatelli
Abstract
The proposed GAN model predicts dose distributions inside unknown phantoms with only small deviations from the full MC simulation with computations times of less than a second. It demonstrates good interpolation ability to unseen but similar phantom geometries and is flexible enough to be trained on data with different radiation scenarios without the need for optimization of the model parameter. This proof-of-concept encourages to apply and further develop the model for the use in MRT treatment planning, which requires fast and accurate predictions with sub-mm resolutions.
Related Papers
- → Divided Voxels: an efficient algorithm for interactive cutting of deformable objects(2020)11 cited
- → Weighted voxel(2018)9 cited
- [Evaluation of therapeutic carbon-beam attenuation in inhomogeneous layered phantoms: Comparison with the present method using a water phantom.].(2006)
- → Surface scan voxel coloring: A surface scan algorithm for scene reconstruction by voxel coloring(2010)
- → Line-Based Structure for Voxel Model(2021)