# Deep Learning based Spectral CT Imaging

arXiv: https://arxiv.org/abs/2008.13570 · HTML: https://mlsanity.com/paper/2008.13570/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Aug 2020.

## Built on by (most cited first, 8 of 8)

- [Hypernetwork-based Personalized Federated Learning for Multi-Institutional CT Imaging](https://mlsanity.com/paper/2206.03709/) (2022)
- [Spectral2Spectral: Image-spectral Similarity Assisted Spectral CT Deep Reconstruction without Reference](https://mlsanity.com/paper/2210.01125/) (2022)
- [Two-and-a-half Order Score-based Model for Solving 3D Ill-posed Inverse Problems](https://mlsanity.com/paper/2308.08511/) (2023)
- [CT Material Decomposition using Spectral Diffusion Posterior Sampling](https://mlsanity.com/paper/2402.03476/) (2024)
- [SpectralCTGaussians: Projection-Domain Reconstruction and Basis Material Decomposition for Spectral CT using 3D Gaussian Splatting](https://mlsanity.com/paper/2609.29638/) (2026)
- [Privacy-Preserving Encrypted Low-Dose CT Denoising](https://mlsanity.com/paper/2310.09101/) (2023)
- [SPLIT: Self-supervised Partitioning for Learned Inversion in Nonlinear Tomography](https://mlsanity.com/paper/2604.15651/) (2026)
- [Shared-Structure 4D Spectral Gaussian Representation for Sparse-View Spectral CT Reconstruction](https://mlsanity.com/paper/2608.16463/) (2026)

## Abstract (arXiv)

Spectral computed tomography (CT) has attracted much attention in radiation dose reduction, metal artifacts removal, tissue quantification and material discrimination. The x-ray energy spectrum is divided into several bins, each energy-bin-specific projection has a low signal-noise-ratio (SNR) than the current-integrating counterpart, which makes image reconstruction a unique challenge. Traditional wisdom is to use prior knowledge based iterative methods. However, this kind of methods demands a great computational cost. Inspired by deep learning, here we first develop a deep learning based reconstruction method; i.e., U-net with Lpp-norm, Total variation, Residual learning, and Anisotropic adaption (ULTRA). Specifically, we emphasize the Various Multi-scale Feature Fusion and Multichannel Filtering Enhancement with a denser connection encoding architecture for residual learning and feature fusion. To address the image deblurring problem associated with the L22-loss, we propose a general Lpp-loss, p>0 Furthermore, the images from different energy bins share similar structures of the same object, the regularization characterizing correlations of different energy bins is incorporated into the Lpp-loss function, which helps unify the deep learning based methods with traditional compressed sensing based methods. Finally, the anisotropically weighted total variation is employed to characterize the sparsity in the spatial-spectral domain to regularize the proposed network. In particular, we validate our ULTRA networks on three large-scale spectral CT datasets, and obtain excellent results relative to the competing algorithms. In conclusion, our quantitative and qualitative results in numerical simulation and preclinical experiments demonstrate that our proposed approach is accurate, efficient and robust for high-quality spectral CT image reconstruction.