Generating Pareto Optimal Dose Distributions for Radiation Therapy Treatment Planning

  • Dan NguyenEmail author
  • Azar Sadeghnejad Barkousaraie
  • Chenyang Shen
  • Xun Jia
  • Steve Jiang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11769)


Radiotherapy treatment planning currently requires many trial-and-error iterations between the planner and treatment planning system, as well as between the planner and physician for discussion/consultation. The physician’s preferences for a particular patient cannot be easily quantified and precisely conveyed to the planner. In this study we present a real-time volumetric Pareto surface dose generation deep learning neural network that can be used after segmentation by the physician, adding a tangible and quantifiable endpoint to portray to the planner. From 70 prostate patients, we first generated 84,000 intensity modulated radiation therapy plans (1,200 plans per patient) sampling the Pareto surface, representing various tradeoffs between the planning target volume (PTV) and the organs-at-risk (OAR), including bladder, rectum, left femur, right femur, and body. We divided the data to 10 test patients and 60 training/validation patients. We then trained a hierarchically densely connected convolutional U-net (HD U-net), to take the PTV and avoidance map representing OARs masks and weights, and predict the optimized plan. The HD U-net is capable of accurately predicting the 3D Pareto optimal dose distributions, with average [mean, max] dose errors of [3.4%, 7.7%](PTV), [1.6%, 5.6%](bladder), [3.7%, 4.2%](rectum), [3.2%, 8.0%](left femur), [2.9%, 7.7%](right femur), and [0.04%, 5.4%](body) of the prescription dose. The PTV dose coverage prediction was also very similar, with errors of 1.3% (D98) and 2.0% (D99). Homogeneity was also similar, differing by 0.06 on average. The neural network can predict the dose within 1.7 s. Clinically, the optimization and dose calculation is much slower, taking 5–10 min.


Radiation therapy treatment planning Intensity modulation Pareto surface Dose distribution Deep learning U-net Neural network 


  1. 1.
    Brahme, A.: Optimization of stationary and moving beam radiation therapy techniques. Radiother. Oncol. 12, 129–140 (1988)CrossRefGoogle Scholar
  2. 2.
    Convery, D., Rosenbloom, M.: The generation of intensity-modulated fields for conformal radiotherapy by dynamic collimation. Phys. Med. Biol. 37, 1359 (1992)CrossRefGoogle Scholar
  3. 3.
    Bortfeld, T.R., Kahler, D.L., Waldron, T.J., Boyer, A.L.: X-ray field compensation with multileaf collimators. Int. J. Radiation Oncol. Biol. Phys. 28, 723–730 (1994)CrossRefGoogle Scholar
  4. 4.
    Yu, C.X.: Intensity-modulated arc therapy with dynamic multileaf collimation: an alternative to tomotherapy. Phys. Med. Biol. 40, 1435 (1995)CrossRefGoogle Scholar
  5. 5.
    Crooks, S.M., Wu, X., Takita, C., Watzich, M., Xing, L.: Aperture modulated arc therapy. Phys. Med. Biol. 48, 1333 (2003)CrossRefGoogle Scholar
  6. 6.
    Earl, M., Shepard, D., Naqvi, S., Li, X., Yu, C.: Inverse planning for intensity-modulated arc therapy using direct aperture optimization. Phys. Med. Biol. 48, 1075 (2003)CrossRefGoogle Scholar
  7. 7.
    Otto, K.: Volumetric modulated arc therapy: IMRT in a single gantry arc. Med. Phys. 35, 310–317 (2008)CrossRefGoogle Scholar
  8. 8.
    Jahn, J.: Scalarization in multi objective optimization. In: Serafini, P. (ed.) Mathematics of Multi Objective Optimization. ICMS, vol. 289, pp. 45–88. Springer, Vienna (1985). Scholar
  9. 9.
    Chambolle, A., Pock, T.: A first-order primal-dual algorithm for convex problems with applications to imaging. J. Math. Imaging Vis. 40, 120–145 (2011)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Nguyen, D., et al.: 3D radiotherapy dose prediction on head and neck cancer patients with a hierarchically densely connected U-net deep learning architecture. Phys. Med. Biol. 64, 065020 (2019)CrossRefGoogle Scholar
  11. 11.
    Huang, G., Liu, Z., van der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: 30th IEEE Conference on Computer Vision and Pattern Recognition, (CVPR 2017), vol. 1, pp. 2261–2269 (2017)Google Scholar
  12. 12.
    Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). Scholar
  13. 13.
    Klambauer, G., Unterthiner, T., Mayr, A., Hochreiter, S.: Self-normalizing neural networks. In: Advances in Neural Information Processing Systems, pp. 971–980. (2017)Google Scholar
  14. 14.
    Grégoire, V., Mackie, T.R.: State of the art on dose prescription, reporting and recording in Intensity-Modulated Radiation Therapy (ICRU report No. 83). Cancer/Radiothérapie 15, 555–559 (2011)CrossRefGoogle Scholar
  15. 15.
    Nguyen, D., et al.: A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning. Sci. Rep. 9, 1076 (2019)CrossRefGoogle Scholar
  16. 16.
    Chen, X., Men, K., Li, Y., Yi, J., Dai, J.: A feasibility study on an automated method to generate patient-specific dose distributions for radiotherapy using deep learning. Med. Phys. 46, 56–64 (2019)CrossRefGoogle Scholar
  17. 17.
    Fan, J., Wang, J., Chen, Z., Hu, C., Zhang, Z., Hu, W.: Automatic treatment planning based on three-dimensional dose distribution predicted from deep learning technique. Med. Phys. 46, 370–381 (2019)CrossRefGoogle Scholar
  18. 18.
    Shiraishi, S., Moore, K.L.: Knowledge-based prediction of three-dimensional dose distributions for external beam radiotherapy. Med. Phys. 43, 378–387 (2016)CrossRefGoogle Scholar
  19. 19.
    Mahmood, R., Babier, A., McNiven, A., Diamant, A., Chan, T.C.: Automated treatment planning in radiation therapy using generative adversarial networks. arXiv preprint arXiv:1807.06489 (2018)
  20. 20.
    Babier, A., Mahmood, R., McNiven, A.L., Diamant, A., Chan, T.C.: Knowledge-based automated planning with three-dimensional generative adversarial networks. arXiv preprint arXiv:1812.09309 (2018)
  21. 21.
    Long, T., Chen, M., Jiang, S.B., Lu, W.: Threshold-driven optimization for reference-based auto-planning. Phys. Med. Biol. 63, 04NT01 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Dan Nguyen
    • 1
    Email author
  • Azar Sadeghnejad Barkousaraie
    • 1
  • Chenyang Shen
    • 1
  • Xun Jia
    • 1
  • Steve Jiang
    • 1
  1. 1.Medical Artificial Intelligence and Automation (MAIA) Laboratory, Department of Radiation OncologyUT Southwestern Medical CenterDallasUSA

Personalised recommendations