# Generating synthetic computed tomography for radiotherapy: SynthRAD2025 challenge report

arXiv: https://arxiv.org/abs/2605.13555 · HTML: https://mlsanity.com/paper/2605.13555/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 52 references that are in the map.
- First posted on arXiv 13 May 2026.

## Builds on (most cited first, 3 of 3)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [TotalSegmentator MRI: Robust Sequence-independent Segmentation of Multiple Anatomic Structures in MRI](https://mlsanity.com/paper/2405.19492/) (2024)

## Built on by (most cited first, 2 of 2)

- [BEAM3R: Beam's-eye-view architecture with Mamba-3 for implicit dose reconstruction](https://mlsanity.com/paper/2609.04747/) (2026)
- [TotalSynth: Robust Whole-Body Synthetic CT from MRI and CBCT](https://mlsanity.com/paper/2609.13838/) (2026)

## Abstract (arXiv)

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±21.3 HU, PSNR ~30 dB, MS-SSIM ~0.936, Dice 0.79, photon γ2%/2mm>98%, proton γ≈85%. Task 2 improved: MAE 48.3±13.4 HU, PSNR 32.6 dB, MS-SSIM 0.968, Dice 0.86, photon γ>99%, proton γ≈89%. Strong image--segmentation correlations (ρ=0.78--0.79) 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.