Deep learning based spectral CT imaging. (December 2021)

• Record Type: Journal Article
• Authors: Wu, Weiwen; Hu, Dianlin; Niu, Chuang; Broeke, Lieza Vanden; Butler, Anthony P.H.; Cao, Peng; Atlas, James; Chernoglazov, Alexander; Vardhanabhuti, Varut; Wang, Ge
• Is Part Of: Neural networks. Volume 144(2021)
• Page Start: 342
• Page End: 358
• Publication Date: 2021-12
• Subjects: Spectral CT -- Deep learning -- Image reconstruction -- Regularization prior -- Lpp loss; Neural computers -- Periodicals; Neural networks (Computer science) -- Periodicals; Neural networks (Neurobiology) -- Periodicals; Nervous System -- Periodicals; Ordinateurs neuronaux -- Périodiques; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux neuronaux (Neurobiologie) -- Périodiques; Neural computers; Neural networks (Computer science); Neural networks (Neurobiology); Periodicals; 006.32
• DOI: 10.1016/j.neunet.2021.08.026
• Languages: English
• ISSNs: 0893-6080

Abstract

Abstract: 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 L p p -norm, T otal variation, R esidual learning, and A nisotropic 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 L 2 2 - loss, we propose a general L p p -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 L p p - 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.