Limited View Tomographic Reconstruction Using a Deep Recurrent Framework with Residual Dense Spatial-Channel Attention Network and Sinogram Consistency

TL;DR

This paper introduces a recurrent deep learning framework with a specialized attention network and sinogram consistency layer to improve limited view tomographic reconstructions, reducing artifacts and noise.

Contribution

It proposes a novel recurrent neural network architecture with residual dense spatial-channel attention and sinogram consistency enforcement for better reconstruction quality.

Findings

Over 5dB PSNR improvement in limited angle reconstruction

Approximately 4dB PSNR gain in sparse view reconstruction

High-quality lesion reconstructions across multiple datasets

Abstract

Limited view tomographic reconstruction aims to reconstruct a tomographic image from a limited number of sinogram or projection views arising from sparse view or limited angle acquisitions that reduce radiation dose or shorten scanning time. However, such a reconstruction suffers from high noise and severe artifacts due to the incompleteness of sinogram. To derive quality reconstruction, previous state-of-the-art methods use UNet-like neural architectures to directly predict the full view reconstruction from limited view data; but these methods leave the deep network architecture issue largely intact and cannot guarantee the consistency between the sinogram of the reconstructed image and the acquired sinogram, leading to a non-ideal reconstruction. In this work, we propose a novel recurrent reconstruction framework that stacks the same block multiple times. The recurrent block consists of a custom-designed residual dense spatial-channel attention network. Further, we develop a sinogram consistency layer interleaved in our recurrent framework in order to ensure that the sampled sinogram is consistent with the sinogram of the intermediate outputs of the recurrent blocks. We evaluate our methods on two datasets. Our experimental results on AAPM Low Dose CT Grand Challenge datasets demonstrate that our algorithm achieves a consistent and significant improvement over the existing state-of-the-art neural methods on both limited angle reconstruction (over 5dB better in terms of PSNR) and sparse view reconstruction (about 4dB better in term of PSNR). In addition, our experimental results on Deep Lesion datasets demonstrate that our method is able to generate high-quality reconstruction for 8 major lesion types.