Enregistré dans:
Détails bibliographiques
Auteurs principaux: Thomsen, Nelas J., Wang, Xinyuan, Lucka, Felix, Demircan-Tureyen, Ezgi
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2602.12755
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917504307167232
author Thomsen, Nelas J.
Wang, Xinyuan
Lucka, Felix
Demircan-Tureyen, Ezgi
author_facet Thomsen, Nelas J.
Wang, Xinyuan
Lucka, Felix
Demircan-Tureyen, Ezgi
contents Diffusion-based image generators are promising priors for ill-posed inverse problems like sparse-view X-ray Computed Tomography (CT). As most studies consider synthetic data, it is not clear whether training data mismatch (``domain shift'') or forward model mismatch complicate their successful application to experimental data. We measured CT data from a physical phantom resembling the synthetic Shepp-Logan phantom and trained diffusion priors on synthetic image data sets with different degrees of domain shift towards it. Then, we employed the priors in a Decomposed Diffusion Sampling scheme on sparse-view CT data sets with increasing difficulty leading to the experimental data. Our results reveal that domain shift plays a nuanced role: while severe mismatch causes model collapse and hallucinations, diverse priors match or exceed well-matched but narrow priors. Forward model mismatch pulls the image samples away from the prior manifold, which causes artifacts but can be mitigated with annealed likelihood weight schedules that also increase computational efficiency. Overall, we demonstrate that performance gains do not immediately translate from synthetic to experimental data, and future development must validate against real-world benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2602_12755
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Towards reconstructing experimental sparse-view X-ray CT data with diffusion models
Thomsen, Nelas J.
Wang, Xinyuan
Lucka, Felix
Demircan-Tureyen, Ezgi
Computer Vision and Pattern Recognition
Diffusion-based image generators are promising priors for ill-posed inverse problems like sparse-view X-ray Computed Tomography (CT). As most studies consider synthetic data, it is not clear whether training data mismatch (``domain shift'') or forward model mismatch complicate their successful application to experimental data. We measured CT data from a physical phantom resembling the synthetic Shepp-Logan phantom and trained diffusion priors on synthetic image data sets with different degrees of domain shift towards it. Then, we employed the priors in a Decomposed Diffusion Sampling scheme on sparse-view CT data sets with increasing difficulty leading to the experimental data. Our results reveal that domain shift plays a nuanced role: while severe mismatch causes model collapse and hallucinations, diverse priors match or exceed well-matched but narrow priors. Forward model mismatch pulls the image samples away from the prior manifold, which causes artifacts but can be mitigated with annealed likelihood weight schedules that also increase computational efficiency. Overall, we demonstrate that performance gains do not immediately translate from synthetic to experimental data, and future development must validate against real-world benchmarks.
title Towards reconstructing experimental sparse-view X-ray CT data with diffusion models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.12755