Abstract
The Gamma index Passing Rate (GPR) is considered the preferred metric to evaluate dose distributions in order to deliver safe radiotherapy treatments. For this reason, in the context of accelerating Monte-Carlo dose simulations using deep neural networks, the GPR remains the default clinical metric used to validate the predictions of the models. However, the optimization criterion that is used for training these neural networks is based on loss functions that are different than GPR. To address this important issue, in this work we introduce a new class of GPR-based loss functions for deep learning. These functions allow us to successfully train neural networks that can directly yield the best dose predictions from a clinical standpoint. Our approach overcomes the mathematical non-differentiability of the GPR, thus allowing a successful application of gradient descent. Moreover, it brings the GPR computation time down to milliseconds, therefore enabling fast trainings. We demonstrate that models trained with our GPR-based loss functions outperform models trained with other commonly used loss functions with respect to several metrics and display a 15% improvement of the GPR over the test data. Code is available at https://rb.gy/vf5jwv.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ahnesjö, A.: Collapsed cone convolution of radiant energy for photon dose calculation in heterogeneous media. Med. Phys. 16(4), 577–592 (1989)
Biggs, S., et al.: Pymedphys: a community effort to develop an open, python-based standard library for medical physics applications. J. Open Source Softw. 7(78), 4555 (2022)
Chen, M., Lu, W., Chen, Q., Ruchala, K., Olivera, G.: Efficient gamma index calculation using fast Euclidean distance transform. Phys. Med. Biol. 54(7), 2037 (2009)
Deng, Y., et al.: Comparison of pencil beam and Monte Carlo calculations with ion chamber array measurements for patient-specific quality assurance. Radiat. Med. Prot. 3(3), 115–122 (2022)
Gu, X., Jia, X., Jiang, S.B.: GPU-based fast gamma index calculation. Phys. Med. Biol. 56(5), 1431 (2011)
Lee, B.I., Boss, M., LaRue, S.M., Martin, T.W., Leary, D.: Comparative study of the collapsed cone convolution and monte carlo algorithms for radiation therapy planning of canine sinonasal tumors reveals significant dosimetric differences. Veterinary Radiol. Ultrasound Official J. Am. Coll. Veterinary Radiol. Int. Veterinary Radiol. Assoc. 63 (2021)
Lin, T.Y., Goyal, P., Girshick, R.B., He, K., Dollár, P.: Focal loss for dense object detection. In: 2017 IEEE ICCV, pp. 2999–3007 (2017)
Liu, Z., Mao, H., Wu, C.Y., Feichtenhofer, C., Darrell, T., Xie, S.: A convnet for the 2020s. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11976–11986 (2022)
Loshchilov, I., Hutter, F.: Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101 (2017)
Martinot, S., Bus, N., Vakalopoulou, M., Robert, C., Deutsch, E., Paragios, N.: High-particle simulation of monte-carlo dose distribution with 3D convlstms. In: MICCAI, pp. 499–508 (2021)
Mohan, R., Chui, C., Lidofsky, L.: Differential pencil beam dose computation model for photons. Med. Phys. 13(1), 64–73 (1986)
Neph, R., Lyu, Q., Huang, Y., Yang, Y.M., Sheng, K.: Deepmc: a deep learning method for efficient monte carlo beamlet dose calculation by predictive denoising in magnetic resonance-guided radiotherapy. Phys. Med. Biol. 66(3), 035022 (2021)
Quan, E., et al.: A comprehensive comparison of IMRT and VMAT plan quality for prostate cancer treatment. Int. J. Radiat. Oncol. Biol. Phys. 83, 1169–78 (2012)
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: MICCAI, pp. 234–241 (2015)
Sudre, C.H., Li, W., Vercauteren, T., Ourselin, S., Jorge Cardoso, M.: Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, pp. 240–248 (2017)
Vasudevan, V., Huang, C., Simiele, E., Yu, L., Xing, L., Schuler, E.: Combining monte carlo with deep learning: Predicting high-resolution, low-noise dose distributions using a generative adversarial network for fast and precise monte carlo simulations. Int. J. Radiat. Oncol. Biol. Phys. 108(3), S44–S45 (2020)
Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
Wendling, M., et al.: A fast algorithm for gamma evaluation in 3D. Med. Phys. 34(5), 1647–1654 (2007)
Zhao, H., Gallo, O., Frosio, I., Kautz, J.: Loss functions for image restoration with neural networks. IEEE Trans. Comput. Imaging 3(1), 47–57 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Martinot, S. et al. (2023). Differentiable Gamma Index-Based Loss Functions: Accelerating Monte-Carlo Radiotherapy Dose Simulation. In: Frangi, A., de Bruijne, M., Wassermann, D., Navab, N. (eds) Information Processing in Medical Imaging. IPMI 2023. Lecture Notes in Computer Science, vol 13939. Springer, Cham. https://doi.org/10.1007/978-3-031-34048-2_37
Download citation
DOI: https://doi.org/10.1007/978-3-031-34048-2_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34047-5
Online ISBN: 978-3-031-34048-2
eBook Packages: Computer ScienceComputer Science (R0)