Unsupervised Self-Prior Embedding Neural Representation for Iterative Sparse-View CT Reconstruction

Emerging unsupervised implicit neural representation (INR) methods, such as NeRP, NeAT, and SCOPE, have shown great potential in addressing sparse-view computed tomography (SVCT) inverse problems. While these INR-based methods perform well on relatively dense SVCT reconstructions, they struggle to achieve comparable performance with supervised methods in sparser SVCT scenarios and are prone to being affected by noise, limiting their applicability in real clinical settings. Additionally, current methods have not fully explored the use of image domain priors for solving SVCT inverse problems. In this work, we demonstrate that imperfect reconstruction results can provide effective image domain priors for INRs to enhance performance. To leverage this, we introduce Self-prior embedding neural representation (Spener), a novel unsupervised method for SVCT reconstruction that integrates iterative reconstruction algorithms. During each iteration, Spener extracts local image prior features from the previous iteration and embeds them to constrain the solution space. Experimental results on multiple CT datasets show that our unsupervised Spener method achieves performance comparable to supervised state-of-the-art (SOTA) methods on in-domain data while outperforming them on out-of-domain datasets. Moreover, Spener significantly improves the performance of INR-based methods in handling SVCT with noisy sinograms.

Authors

• Xuanyu Tian ShanghaiTech University Linggang Laboratory
• Lixuan Chen University of Michigan - Ann Arbor
• Qing Wu ShanghaiTech University
• Chenhe Du ShanghaiTech University
• Jingjing Shi Huazhong University of Science and Technology
• Hongjiang Wei School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
• Yuyao Zhang School of Information Science and Technology, ShanghaiTech University, Shanghai, China

Keywords

No keywords are indexed for this paper.