aiqbench analysis
This research introduces EqDiff-CT, an equivariant conditional diffusion model designed to synthesize high-fidelity CT images from CBCT scans for adaptive radiotherapy. By enforcing rotational symmetry through steerable layers within its U-Net backbone, the method effectively mitigates artifacts and improves Hounsfield Unit accuracy critical for dose calculations. The approach represents a significant advancement in medical imaging workflows where AI-driven generative models bridge gaps between real-time low-dose modalities and high-quality planning scans.
Equivariant Conditional Diffusion Model for Head and Neck CT Image Synthesis from CBCT
arXiv:2509.21913v2 Announce Type: replace-cross Abstract: Background: Cone-beam computed tomography CBCT is a commonly used modality for image guided radiotherapy. It offers real time anatomical visualization with low acquisition cost and dose. Nevertheless, photon scattering and beam hindrance lead CBCT images to suffer from several artifacts. These involve inaccurate Hounsfield Unit HU values, which render a lower reliability towards the dose calculations and adaptive planning. Purpose: Computed tomography CT, on the contrary, offers better image quality and accurate HU calibration, yet is typically acquired using offline mode and fails to capture the intra-treatment anatomical changes. This renders a need for developing an accurate CBCT to CT synthesis to mitigate the gap in imaging quality in the adaptive radiotherapy workflow. Methods: We propose a novel diffusion based conditional generative model, coined EqDiff-CT, to synthesize high quality CT images from CBCT. EqDiff-CT employs a denoising diffusion probabilistic model DDPM to iteratively inject noise and learn latent representations that enable reconstruction of anatomically consistent CT images. A group equivariant conditional U-Net backbone, implemented with e2cnn steerable layers, enforces rotational equivariance cyclic C4 symmetry, helping preserve fine structural details while minimizing noise and artifacts. Results: The system was trained and validated on the SynthRAD2025 dataset, comprising CBCT-CT scans across multiple head and neck anatomical sites, and we compared it with advanced methods such as CycleGAN and DDPM. EqDiff-CT provided substantial gains in structural fidelity, HU accuracy and quantitative metrics. Visual findings confirm the improved recovery, sharper soft tissue boundaries, and realistic bone reconstructions. Conclusions: The findings suggest that the diffusion model has offered a robust and generalizable framework for CBCT improvements.
Abstract:Background: Cone-beam computed tomography CBCT is a commonly used modality for image guided radiotherapy. It offers real time anatomical visualization with low acquisition cost and dose. Nevertheless, photon scattering and beam hindrance lead CBCT images to suffer from several artifacts. These involve inaccurate Hounsfield Unit HU values, which render a lower reliability towards the dose calculations and adaptive planning. Purpose: Computed tomography CT, on the contrary, offers better image quality and accurate HU calibration, yet is typically acquired using offline mode and fails to capture the intra-treatment anatomical changes. This renders a need for developing an accurate CBCT to CT synthesis to mitigate the gap in imaging quality in the adaptive radiotherapy workflow. Methods: We propose a novel diffusion based conditional generative model, coined EqDiff-CT, to synthesize high quality CT images from CBCT. EqDiff-CT employs a denoising diffusion probabilistic model DDPM to iteratively inject noise and learn latent representations that enable reconstruction of anatomically consistent CT images. A group equivariant conditional U-Net backbone, implemented with e2cnn steerable layers, enforces rotational equivariance cyclic C4 symmetry, helping preserve fine structural details while minimizing noise and artifacts. Results: The system was trained and validated on the SynthRAD2025 dataset, comprising CBCT-CT scans across multiple head and neck anatomical sites, and we compared it with advanced methods such as CycleGAN and DDPM. EqDiff-CT provided substantial gains in structural fidelity, HU accuracy and quantitative metrics. Visual findings confirm the improved recovery, sharper soft tissue boundaries, and realistic bone reconstructions. Conclusions: The findings suggest that the diffusion model has offered a robust and generalizable framework for CBCT improvements.
Medical Physics (physics.med-ph); Artificial Intelligence (cs.AI)
arXiv:2509.21913 [physics.med-ph]
(or arXiv:2509.21913v2 [physics.med-ph] for this version)
https://doi.org/10.48550/arXiv.2509.21913 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1002/mp.70617 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 366 words
Related coverage
EPC-3D-Diff: Equivariant Physics Consistent Conditional 3D Latent Diffusion for CBCT to CT Synthesis
arXiv · Sep 9, 2026
Let It Be Simple: One-Step Action Generation for Vision-Language-Action Models
arXiv · Jun 6, 2026
Multimodal Co-Generation With Continuous Bitstream Diffusion
arXiv · Oct 14, 2026
Freeze, Share, Shrink: Rethinking the Action Backbone in Diffusion Policies
arXiv · Sep 16, 2026