Geometric Constraints Enable Self-Supervised Sinogram Inpainting in Sparse-View Tomography

TL;DR

This paper introduces a self-supervised method for inpainting missing projections in sparse-view CT scans, leveraging geometric constraints to improve image quality without large training datasets.

Contribution

The work presents a novel self-supervised projection inpainting technique that directly optimizes missing views using geometric constraints, applicable to various tomographic imaging scenarios.

Findings

Improves reconstruction quality by up to 17.6% in SSIM.

Effective on real X-ray microscope tomographic data.

Operates without large training datasets.

Abstract

The diagnostic quality of computed tomography (CT) scans is usually restricted by the induced patient dose, scan speed, and image quality. Sparse-angle tomographic scans reduce radiation exposure and accelerate data acquisition, but suffer from image artifacts and noise. Existing image processing algorithms can restore CT reconstruction quality but often require large training data sets or can not be used for truncated objects. This work presents a self-supervised projection inpainting method that allows optimizing missing projective views via gradient-based optimization. By reconstructing independent stacks of projection data, a self-supervised loss is calculated in the CT image domain and used to directly optimize projection image intensities to match the missing tomographic views constrained by the projection geometry. Our experiments on real X-ray microscope (XRM) tomographic mouse tibia bone scans show that our method improves reconstructions by 3.1-7.4%/7.7-17.6% in terms of PSNR/SSIM with respect to the interpolation baseline. Our approach is applicable as a flexible self-supervised projection inpainting tool for tomographic applications.