Resolution-Independent Neural Operators for Multi-Rate Sparse-View CT

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Datta, Aujasvit, Wang, Jiayun, Aali, Asad, Jatyani, Armeet Singh, Anandkumar, Anima
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909959752515584
author Datta, Aujasvit
Wang, Jiayun
Aali, Asad
Jatyani, Armeet Singh
Anandkumar, Anima
author_facet Datta, Aujasvit
Wang, Jiayun
Aali, Asad
Jatyani, Armeet Singh
Anandkumar, Anima
contents Sparse-view Computed Tomography (CT) reconstructs images from a limited number of X-ray projections to reduce radiation and scanning time, which makes reconstruction an ill-posed inverse problem. Deep learning methods achieve high-fidelity reconstructions but often overfit to a fixed acquisition setup, failing to generalize across sampling rates and image resolutions. For example, convolutional neural networks (CNNs) use the same learned kernels across resolutions, leading to artifacts when data resolution changes. We propose Computed Tomography neural Operator (CTO), a unified CT reconstruction framework that extends to continuous function space, enabling generalization (without retraining) across sampling rates and image resolutions. CTO operates jointly in the sinogram and image domains through rotation-equivariant Discrete-Continuous convolutions parametrized in the function space, making it inherently resolution- and sampling-agnostic. Empirically, CTO enables consistent multi-sampling-rate and cross-resolution performance, with on average >4dB PSNR gain over CNNs. Compared to state-of-the-art diffusion methods, CTO is 500$\times$ faster in inference time with on average 3dB gain. Empirical results also validate our design choices behind CTO's sinogram-space operator learning and rotation-equivariant convolution. Overall, CTO outperforms state-of-the-art baselines across sampling rates and resolutions, offering a scalable and generalizable solution that makes automated CT reconstruction more practical for deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2512_12236
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Resolution-Independent Neural Operators for Multi-Rate Sparse-View CT
Datta, Aujasvit
Wang, Jiayun
Aali, Asad
Jatyani, Armeet Singh
Anandkumar, Anima
Image and Video Processing
Computer Vision and Pattern Recognition
Sparse-view Computed Tomography (CT) reconstructs images from a limited number of X-ray projections to reduce radiation and scanning time, which makes reconstruction an ill-posed inverse problem. Deep learning methods achieve high-fidelity reconstructions but often overfit to a fixed acquisition setup, failing to generalize across sampling rates and image resolutions. For example, convolutional neural networks (CNNs) use the same learned kernels across resolutions, leading to artifacts when data resolution changes. We propose Computed Tomography neural Operator (CTO), a unified CT reconstruction framework that extends to continuous function space, enabling generalization (without retraining) across sampling rates and image resolutions. CTO operates jointly in the sinogram and image domains through rotation-equivariant Discrete-Continuous convolutions parametrized in the function space, making it inherently resolution- and sampling-agnostic. Empirically, CTO enables consistent multi-sampling-rate and cross-resolution performance, with on average >4dB PSNR gain over CNNs. Compared to state-of-the-art diffusion methods, CTO is 500$\times$ faster in inference time with on average 3dB gain. Empirical results also validate our design choices behind CTO's sinogram-space operator learning and rotation-equivariant convolution. Overall, CTO outperforms state-of-the-art baselines across sampling rates and resolutions, offering a scalable and generalizable solution that makes automated CT reconstruction more practical for deployment.
title Resolution-Independent Neural Operators for Multi-Rate Sparse-View CT
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.12236