A research team led by Professor WANG Hongzhi from the Institute of Health and Medical Technology, the Hefei Institute of Physical Science of the Chinese Academy of Sciences, has developed a multi-stage dual-domain progressive network with synergistic training for sparse-view CT (SVCT) reconstruction.
The related study was published in Neural Networks recently.
Sparse-view CT aims to reduce patient radiation exposure and shorten scanning time by decreasing the number of projection angles. However, this reduction often introduces severe streak artifacts, compromising diagnostic reliability. Conventional deep learning-based methods typically require separate models for different view conditions, resulting in inefficient workflows and limited adaptability.
In this study, the researchers proposed a synergistic multi-stage dual-domain progressive reconstruction framework (MDPRNet) that introduces two key innovations.
First, a multi-view synergistic training strategy groups the data into ultra-sparse and sparse views, allowing a single unified model to adapt across a wide range of sampling conditions. This strategy effectively reduces performance discrepancies between view intervals and ensures stability under extremely sparse scenarios.
Second, a multi-stage dual-domain progressive architecture combines features from both sinogram and image domains, while a Cross-stage Feature Adapter with attention modules enhances feature fusion and progressively improves reconstruction quality.
Validated on both public and self-built CT datasets, MDPRNet consistently outperformed existing methods and maintained high reconstruction accuracy and robustness under all sparse-view conditions.
"Our model solves the problem of adapting to different sparse-view settings,”said Prof. WANG, "It greatly improves reconstruction accuracy and generalization."
Overall architecture of the proposed MDPRNet model for sparse-view CT reconstruction, employing a multi-stage progressive reconstruction framework. （Image by SHAO Jingyuan）