# Dual Network Architecture for Few-view CT -- Trained on ImageNet Data and Transferred for Medical Imaging

## TL;DR

This paper introduces a dual network architecture (DNA) for few-view CT image reconstruction that reduces radiation dose and can be pre-trained on natural images, demonstrating competitive performance with less memory usage.

## Contribution

The paper presents a novel dual network architecture for sinogram-based CT reconstruction that uses fewer parameters and can be pre-trained on natural images to improve performance.

## Key findings

- DNA achieves competitive results compared to state-of-the-art methods.
- Pre-training on natural images helps prevent overfitting with limited patient data.
- The method significantly reduces memory requirements during training.

## Abstract

X-ray computed tomography (CT) reconstructs cross-sectional images from projection data. However, ionizing X-ray radiation associated with CT scanning might induce cancer and genetic damage. Therefore, the reduction of radiation dose has attracted major attention. Few-view CT image reconstruction is an important topic to reduce the radiation dose. Recently, data-driven algorithms have shown great potential to solve the few-view CT problem. In this paper, we develop a dual network architecture (DNA) for reconstructing images directly from sinograms. In the proposed DNA method, a point-based fully-connected layer learns the backprojection process requesting significantly less memory than the prior arts do. Proposed method uses O(C*N*N_c) parameters where N and N_c denote the dimension of reconstructed images and number of projections respectively. C is an adjustable parameter that can be set as low as 1. Our experimental results demonstrate that DNA produces a competitive performance over the other state-of-the-art methods. Interestingly, natural images can be used to pre-train DNA to avoid overfitting when the amount of real patient images is limited.