English

Generating synthetic computed tomography for radiotherapy: SynthRAD2025 challenge report

Medical Physics 2026-05-14 v1 Artificial Intelligence

Abstract

Radiation therapy (RT) requires precise dose delivery over multiple fractions, with CT fundamental for treatment planning due to its electron density information. Repeated CT acquisitions impose radiation exposure and logistical burdens, MRI lacks electron density, and cone-beam CT (CBCT) requires correction for dose calculation. Synthetic CT (sCT) generation addresses these by converting MRI or CBCT into CT-equivalent images with accurate Hounsfield Unit (HU) values, enabling MRI-only RT and CBCT-based adaptive workflows. Building on SynthRAD2023, SynthRAD2025 benchmarked sCT methods on 2,362 patients from five European centers across head and neck, thorax, and abdomen. Two tasks: MRI-to-CT (890 cases) and CBCT-to-CT (1,472 cases), evaluated via image similarity (MAE, PSNR, MS-SSIM), segmentation (Dice, HD95), and dosimetric metrics from photon and proton plans. With 803 participants and 12/13 valid submissions, Task 1 top performance reached MAE 64.8±21.364.8\pm21.3 HU, PSNR \sim30 dB, MS-SSIM \sim0.936, Dice 0.79, photon γ2%/2mm>98%\gamma_{2\%/2\text{mm}}>98\%, proton γ85%\gamma\approx85\%. Task 2 improved: MAE 48.3±13.448.3\pm13.4 HU, PSNR 32.6 dB, MS-SSIM 0.968, Dice 0.86, photon γ>99%\gamma>99\%, proton γ89%\gamma\approx89\%. Strong image--segmentation correlations (ρ=0.78\rho=0.78--0.790.79) but moderate dose correlations confirmed image quality is insufficient as a dosimetric surrogate. Head-and-neck cases were most consistent; thoracic and abdominal cases showed greater variability. Residual errors at tissue interfaces propagate along beam paths, affecting proton dose more than photon. SynthRAD2025 demonstrates that deep learning yields clinically relevant sCTs, especially for CBCT-to-CT, while identifying persistent MRI-to-CT challenges and underscoring dose-based evaluation as essential for clinical validation.

Keywords

Cite

@article{arxiv.2605.13555,
  title  = {Generating synthetic computed tomography for radiotherapy: SynthRAD2025 challenge report},
  author = {Viktor Rogowski and Maarten L. Terpstra and Niklas Wahl and Florian Kamp and Erik van der Bijl and Arthur Jr. Galapon and Christopher Kurz and Bowen Xin and Zhengxiang Sun and Hollie Min and Gregg Belous and Jason Dowling and Yan Xia and Siyuan Mei and Fuxin Fan and Arthur Longuefosse and Javier Sequeiro Gonzalez and Miguel Diaz Benito and Alvaro Garcia Martin and Fabien Baldacci and Valentin Boussot and Cédric Hémon and Jean-Claude Nunes and Jean-Louis Dillenseger and Zhiyuan Zhang and Jinghua Cai and Han Bing and Tan Zuopeng and Ricardo Brioso and Daniele Loiacono and Guillaume Landry and Adrian Thummerer and Matteo Maspero},
  journal= {arXiv preprint arXiv:2605.13555},
  year   = {2026}
}

Comments

59 pages total: 26 pages main article + supplementary material; 8 figures in the main manuscript and 3 supplementary figures. Currently under review at the journal Medical Image Analysis (MIA)

R2 v1 2026-07-22T07:10:12.231Z