ResDynUNet++: A nested U-Net with residual dynamic convolution blocks for dual-spectral CT

TL;DR

This paper introduces ResDynUNet++, a novel neural network architecture that enhances dual-spectral CT reconstruction by combining knowledge-driven and data-driven methods, leading to improved image quality and artifact reduction.

Contribution

The paper presents a hybrid framework integrating iterative reconstruction with a new residual dynamic convolution UNet++ for better dual-spectral CT images.

Findings

Outperforms existing methods on synthetic and clinical datasets.

Effectively reduces artifacts and improves image accuracy.

Demonstrates fast convergence and stable training.

Abstract

We propose a hybrid reconstruction framework for dual-spectral CT (DSCT) that integrates iterative methods with deep learning models. The reconstruction process consists of two complementary components: a knowledge-driven module and a data-driven module. In the knowledge-driven phase, we employ the oblique projection modification technique (OPMT) to reconstruct an intermediate solution of the basis material images from the projection data. We select OPMT for this role because of its fast convergence, which allows it to rapidly generate an intermediate solution that successfully achieves basis material decomposition. Subsequently, in the data-driven phase, we introduce a novel neural network, ResDynUNet++, to refine this intermediate solution. The ResDynUNet++ is built upon a UNet++ backbone by replacing standard convolutions with residual dynamic convolution blocks, which combine the adaptive, input-specific feature extraction of dynamic convolution with the stable training of residual connections. This architecture is designed to address challenges like channel imbalance and near-interface large artifacts in DSCT, producing clean and accurate final solutions. Extensive experiments on both synthetic phantoms and real clinical datasets validate the efficacy and superior performance of the proposed method.