Multi-stage Deep Learning Artifact Reduction for Pallel-beam Computed Tomography

TL;DR

This paper presents a multi-stage deep learning approach for artifact reduction in synchrotron computed tomography, improving image quality by addressing artifacts at each pipeline stage with specialized models.

Contribution

The authors introduce a novel multi-stage deep learning pipeline with bypass connections for artifact reduction in synchrotron CT, enhancing effectiveness and efficiency over existing methods.

Findings

Effective artifact reduction demonstrated on simulated data

Outperforms existing methods in real-world datasets

Minimizes error propagation through bypass connections

Abstract

Computed Tomography (CT) using synchrotron radiation is a powerful technique that, compared to lab-CT techniques, boosts high spatial and temporal resolution while also providing access to a range of contrast-formation mechanisms. The acquired projection data is typically processed by a computational pipeline composed of multiple stages. Artifacts introduced during data acquisition can propagate through the pipeline, and degrade image quality in the reconstructed images. Recently, deep learning has shown significant promise in enhancing image quality for images representing scientific data. This success has driven increasing adoption of deep learning techniques in CT imaging. Various approaches have been proposed to incorporate deep learning into computational pipelines, but each has limitations in addressing artifacts effectively and efficiently in synchrotron CT, either in properly addressing the specific artifacts, or in computational efficiency. Recognizing these challenges, we introduce a novel method that incorporates separate deep learning models at each stage of the tomography pipeline-projection, sinogram, and reconstruction-to address specific artifacts locally in a data-driven way. Our approach includes bypass connections that feed both the outputs from previous stages and raw data to subsequent stages, minimizing the risk of error propagation. Extensive evaluations on both simulated and real-world datasets illustrate that our approach effectively reduces artifacts and outperforms comparison methods.