ESTRO 2024 - Abstract Book

S3765

Physics - Image acquisition and processing

ESTRO 2024

photon

and

proton

plan,

the

correlation

was

0.79-0.8).

Figure 2 Overview of the top-ranked submissions for a brain (left) and pelvic (right) case showing the sCT (first row), ground truth CT (first column, top row), difference sCT minus CT (middle row), and photon (left), proton (right) dose plan with dose error (CT - sCT).

Conclusion:

The SynthRAD2023 Grand Challenge provided a platform for researchers to investigate and benchmark sCT generation techniques, offering insights into developing algorithms for MRI-only and CBCT-based adaptive radiotherapy. Image similarity cannot be considered a substitute for clinical dose evaluation to assess the accuracy of the sCT solution.

Keywords: synthetic CT, deep learning, MRI and CBCT

References:

1 Spadea MF & Maspero M, Zaffino P, Seco J. (2021). Deep learning-based synthetic-CT generation in radiotherapy and PET: A review. Medical Physics, 48(11), 6537–6566. https://doi.org/10.1002/mp.15150

2 Thummerer A, van der Bijl E, Galapon Jr A, Verhoeff JJ, Langendijk JA, Both S, van den Berg CAT, Maspero M. (2023). SynthRAD2023 Grand Challenge dataset: Generating synthetic CT for radiotherapy. Med Phys 2023;1-11. https://doi.org/10.1002/mp.16529 3 Thummerer, Adrian, van der Bijl, Erik, & Maspero, Matteo. (2023). SynthRAD2023 Grand Challenge dataset: synthesizing computed tomography for radiotherapy (0.1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.7260705