aiqbench analysis
This research introduces Dose-Aware Cold Diffusion (DACD), a novel framework that integrates physics-based consistency into diffusion models to reconstruct medical images from ultra-low-dose CT scans. By treating radiation dose as a continuous latent variable, the method achieves robust generalization across unseen dose levels, outperforming current state-of-the-art techniques in both quantitative accuracy and visual fidelity. The approach represents a significant advancement in applying generative AI to solve critical clinical challenges where data scarcity and safety constraints are paramount.
Dose-Aware Cold Diffusion with Physics Consistency for Generalizable Low-Dose CT Reconstruction
arXiv:2609.18943v1 Announce Type: cross Abstract: Reducing radiation dose in computed tomography significantly degrades image quality and poses challenges for accurate and clinically reliable reconstruction. While recent approaches have shown promise for low-dose CT, they often struggle to generalize across continuous and previously unseen dose levels, leading to artifacts and loss of anatomical detail. To address these limitations, we propose Dose-Aware Cold Diffusion (DACD), a physics-consistent reconstruction framework that explicitly models radiation dose as a continuous latent factor within a cold diffusion process. The proposed DACD framework integrates image-based dose-aware perception, multi-scale structural prior extraction, and dose-calibrated step allocation to adaptively guide the denoising trajectory. In addition, an iterative forward-backprojection correction is incorporated into the reverse refinement process to enforce projection-domain data consistency. Extensive experiments on three public benchmarks, including Mayo-2020, Mayo-2016, and LoDoPaB-CT, demonstrate that DACD consistently outperforms state-of-the-art diffusion-based and physics-guided methods in both quantitative accuracy and visual fidelity, particularly under ultra-low-dose conditions. The results show that DACD achieves robust generalization across a continuous range of dose levels, including those unseen during training.
Abstract:Reducing radiation dose in computed tomography significantly degrades image quality and poses challenges for accurate and clinically reliable reconstruction. While recent approaches have shown promise for low-dose CT, they often struggle to generalize across continuous and previously unseen dose levels, leading to artifacts and loss of anatomical detail. To address these limitations, we propose Dose-Aware Cold Diffusion (DACD), a physics-consistent reconstruction framework that explicitly models radiation dose as a continuous latent factor within a cold diffusion process. The proposed DACD framework integrates image-based dose-aware perception, multi-scale structural prior extraction, and dose-calibrated step allocation to adaptively guide the denoising trajectory. In addition, an iterative forward-backprojection correction is incorporated into the reverse refinement process to enforce projection-domain data consistency. Extensive experiments on three public benchmarks, including Mayo-2020, Mayo-2016, and LoDoPaB-CT, demonstrate that DACD consistently outperforms state-of-the-art diffusion-based and physics-guided methods in both quantitative accuracy and visual fidelity, particularly under ultra-low-dose conditions. The results show that DACD achieves robust generalization across a continuous range of dose levels, including those unseen during training.
8 pages, 7 figures, 3 tables. Accepted at International Joint Conference on Neural Networks (IJCNN 2026)
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
I.4.4; I.2.10; J.3
arXiv:2609.18943 [cs.CV]
(or arXiv:2609.18943v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.18943 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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 · 288 words
Related coverage
Links #5: 2026/07
LessWrong · Jul 31, 2026
Equivariant Conditional Diffusion Model for Head and Neck CT Image Synthesis from CBCT
arXiv · Jul 24, 2026
Models Can't Remember Their Training. Neither Can You.
LessWrong · Jul 19, 2026
Current AIs out-persuade professionals in lab settings but (probably) not in the world
LessWrong · Oct 5, 2026