TomoSelfDEQ: Self-Supervised Deep Equilibrium Learning for Sparse-Angle CT Reconstruction

Tatiana A. Bubba, Matteo Santacesaria, Andrea Sebastiani
2/28/2025
eess.IVcs.CV

No official code found. Best implementation: sedaboni/TomoSelfDEQ (4 stars, last commit 2025-08-20).

Abstract

Deep learning has emerged as a powerful tool for solving inverse problems in imaging, including computed tomography (CT). However, most approaches require paired training data with ground truth images, which can be difficult to obtain, e.g., in medical applications. We present TomoSelfDEQ, a self-supervised Deep Equilibrium (DEQ) framework for sparse-angle CT reconstruction that trains directly on undersampled measurements. We establish theoretical guarantees showing that, under suitable assumptions, our self-supervised updates match those of fully-supervised training with a loss including the (possibly non-unitary) forward operator like the CT forward map. Numerical experiments on sparse-angle CT data confirm this finding, also demonstrating that TomoSelfDEQ outperforms existing self-supervised methods, achieving state-of-the-art results with as few as 16 projection angles.

View on arXivView PDF

Code Implementations(1)

sedaboni/TomoSelfDEQCommunity100%

TomoSelfDEQ: Self-Supervised Deep Equilibrium Learning for Sparse-Angle CT Reconstruction (2025)

40PythonSep 30, 20241 years ago
deep-equilibrium-modelsimage-restorationself-supervisedtomography

Similar Papers

Equivariant Splitting: Self-supervised learning from incomplete data

85% match1 repo42025

DPER: Diffusion Prior Driven Neural Representation for Limited Angle and Sparse View CT Reconstruction

84% match2024

Benchmarking Self-Supervised Learning Methods for Accelerated MRI Reconstruction

83% match3 repos42025

Swap-Net: A Memory-Efficient 2.5D Network for Sparse-View 3D Cone Beam CT Reconstruction

82% match2024

The method of the approximate inverse for limited-angle CT

82% match2025

Coordinate-based Speed of Sound Recovery for Aberration-Corrected Photoacoustic Computed Tomography

82% match1 repo162024

Cite this paper

@article{bubba2025tomoselfdeq,
  title  = {TomoSelfDEQ: Self-Supervised Deep Equilibrium Learning for Sparse-Angle CT Reconstruction},
  author = {Tatiana A. Bubba and Matteo Santacesaria and Andrea Sebastiani},
  year   = {2025},
  eprint = {2502.21320},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2502.21320v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2502.21320.svg)](https://codeofpaper.com/paper/2502.21320)

Discussion