Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction
This work addresses the problem of efficient and accurate CT image reconstruction for medical imaging, though it appears incremental as it builds on variational models with deep learning enhancements.
The paper tackles sparse-view CT reconstruction by proposing a Learned Alternating Minimization Algorithm (LAMA) that uses learnable nonsmooth nonconvex regularizers in both image and sinogram domains, resulting in improved reconstruction accuracy and outperforming existing methods on benchmark datasets.
We propose a novel Learned Alternating Minimization Algorithm (LAMA) for dual-domain sparse-view CT image reconstruction. LAMA is naturally induced by a variational model for CT reconstruction with learnable nonsmooth nonconvex regularizers, which are parameterized as composite functions of deep networks in both image and sinogram domains. To minimize the objective of the model, we incorporate the smoothing technique and residual learning architecture into the design of LAMA. We show that LAMA substantially reduces network complexity, improves memory efficiency and reconstruction accuracy, and is provably convergent for reliable reconstructions. Extensive numerical experiments demonstrate that LAMA outperforms existing methods by a wide margin on multiple benchmark CT datasets.