Unsupervised Deep Learning for Inverse Problems in Computed Tomography

arXiv:2508.05321 · physics.med-ph, cs.AI · Submitted 2025-08-07 · Read on arXiv

physics.med-ph, cs.AI

Submitted: 2025-08-07

Updated: 2026-07-17

Code: https://github.com/gschramm/parallelproj

Terminology

• Inverse Problem — 3× in this paper · explained in 2 episodes
• Adaptability — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Deep Image Prior — 1× in this paper
• Over-smoothing — 1× in this paper · explained in 4 episodes
• Uncertainty Quantification — 1× in this paper · explained in 9 episodes

Sources

• Deep Image Prior
• LEARN++: Recurrent Dual-Domain Reconstruction Network for Compressed Sensing CT
• Deep Radon Prior: A Fully Unsupervised Framework for Sparse-View CT Reconstruction
• Equivariant Imaging: Learning Beyond the Range Space
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Aggregated Residual Transformations for Deep Neural Networks
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• Deep Residual Learning for Image Recognition
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Related papers

• Reduced NEXI protocol for the quantification of human gray matter microstructure on the Connectome 2.0 scanner
• Edge-Aligned Beam Placement in Scanning Probe Tomography via Reconstruction-Free Sequential Design of Experiments
• Validation of a Computational Respiratory System Model for Mechanical Ventilation
• Cross-View Attention Fusion Net: A Prior-Guided Dual-View Representation Learning for Cardiac Output Estimation from Short-Term PPG Signals