ESTRO 2024 - Abstract Book

S3776

Physics - Image acquisition and processing

ESTRO 2024

Diffusion Schrödinger Bridge Models (DSBM) present a significant stride forward in the MR to CT synthesis domain, addressing the long-standing challenge caused by data misalignment and registration errors. By grounding our approach in the principles of the Schrödinger Bridge and capitalizing on the inherent structural information of MR images, we achieved not only a reduction in synthesis errors but also a notable enhancement in geometric preservation. The promising results from our DSBM approach, combined with its potential applications in on-board imaging guidance, underscore its clinical significance and set a new benchmark for future endeavors in MR-based CT synthesis.

Keywords: MR-CT Synthesis, Diffusion Models, Proton Therapy

References:

[1] Boulanger, M., et al. "Deep learning methods to generate synthetic CT from MRI in radiotherapy: A literature review." Physica Medica 89 (2021): 265-281.

[2] Schrödinger, Erwin. "Sur la théorie relativiste de l'électron et l'interprétation de la mécanique quantique." Annales de l'institut Henri Poincaré. Vol. 2. No. 4. 1932.

[3] Han, Xiao. "MR ‐ based synthetic CT generation using a deep convolutional neural network method." Medical physics 44.4 (2017): 1408-1419.

[4] Ho, Jonathan, Ajay Jain, and Pieter Abbeel. "Denoising diffusion probabilistic models." Advances in neural information processing systems 33 (2020): 6840-6851.

[5] Song, Yang, et al. "Score-based generative modeling through stochastic differential equations." arXiv preprint arXiv:2011.13456 (2020).

[6] Çiçek, Özgün, et al. "3D U-Net: learning dense volumetric segmentation from sparse annotation." Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016: 19th International Conference, Athens, Greece, October 17-21, 2016, Proceedings, Part II 19. Springer International Publishing, 2016.

[7] Cohen, Israel, et al. "Pearson correlation coefficient." Noise reduction in speech processing (2009): 1-4.

552

Mini-Oral

First-in-human breast imaging study with ultra-low field MRI for compact proton therapy systems

Friderike K. Longarino 1,2,3 , Sheng Shen 4,5 , Neha Koonjoo 4,5 , Susu Yan 1,6 , Rachel B. Jimenez 1,6 , Matthew S. Rosen 4,5 , Thomas R. Bortfeld 1,6 1 Massachusetts General Hospital, Department of Radiation Oncology, Boston, USA. 2 German Cancer Research Center (DKFZ), Clinical Cooperation Unit Translational Radiation Oncology, Heidelberg, Germany. 3 Heidelberg University