Corpus ID: 204509240

# Extreme Few-view CT Reconstruction using Deep Inference

@article{Kim2019ExtremeFC,
title={Extreme Few-view CT Reconstruction using Deep Inference},
author={Hyojin Kim and Rushil Anirudh and K. Aditya Mohan and Kyle Champley},
journal={ArXiv},
year={2019},
volume={abs/1910.05375}
}
Reconstruction of few-view x-ray Computed Tomography (CT) data is a highly ill-posed problem. It is often used in applications that require low radiation dose in clinical CT, rapid industrial scanning, or fixed-gantry CT. Existing analytic or iterative algorithms generally produce poorly reconstructed images, severely deteriorated by artifacts and noise, especially when the number of x-ray projections is considerably low. This paper presents a deep network-driven approach to address extreme few… Expand
2 Citations

#### Figures and Topics from this paper

Dynamic CT Reconstruction from Limited Views with Implicit Neural Representations and Parametric Motion Fields
Reconstructing dynamic, time-varying scenes with computed tomography (4D-CT) is a challenging and ill-posed problem common to industrial and medical settings. Existing 4D-CT reconstructions areExpand
A Survey of Soft Computing Approaches in Biomedical Imaging
• Computer Science, Medicine
• Journal of healthcare engineering
• 2021
A short review of soft computing approaches such as fuzzy logic, artificial neural network, genetic algorithm, machine learning, and deep learning for medical imaging modalities and the possible research strategies for further development are proposed. Expand

#### References

SHOWING 1-10 OF 20 REFERENCES
DEEP BACK PROJECTION FOR SPARSE-VIEW CT RECONSTRUCTION
• Computer Science, Engineering
• 2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP)
• 2018
This paper uses a deep convolutional neural network to produce high-quality reconstructions directly from sinogram data and demonstrates the benefit of the CNN based back projection on simulated sparse-view CT data over classical FBP. Expand
Low-Dose X-ray CT Reconstruction via Dictionary Learning
• Computer Science, Medicine
• IEEE Transactions on Medical Imaging
• 2012
The results show that the proposed approach might produce better images with lower noise and more detailed structural features in the authors' selected cases, however, there is no proof that this is true for all kinds of structures. Expand
Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion
• Computer Science, Mathematics
• 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
• 2018
This paper proposes to address the problem of CT reconstruction using CTNet - a system of 1D and 2D convolutional neural networks that operates directly on a limited angle sinogram to predict the reconstruction, and proposes a measure of confidence for the reconstruction that enables a practitioner to gauge the reliability of a prediction made by CTNet. Expand
CNN-Based Projected Gradient Descent for Consistent CT Image Reconstruction
• Computer Science, Mathematics
• IEEE Transactions on Medical Imaging
• 2018
A relaxed version of PGD wherein gradient descent enforces measurement consistency, while a CNN recursively projects the solution closer to the space of desired reconstruction images and shows an improvement over total variation-based regularization, dictionary learning, and a state-of-the-art deep learning-based direct reconstruction technique. Expand
Improved total variation-based CT image reconstruction applied to clinical data.
• Mathematics, Medicine
• Physics in medicine and biology
• 2011
A new method to adapt the step-size adaption of the ASD-POCS algorithm to solve the problems of angular undersampling, data lost due to metal implants, limited view angle tomography and interior tomography. Expand
On Image Reconstruction from a Small Number of Projections.
• Mathematics, Computer Science
• Inverse problems
• 2008
It is demonstrated that sometimes an algorithm based on total variation minimization produces medically-desirable reconstructions in computerized tomography (CT) even from a small number of projections, and that such a reconstruction is not guaranteed to provide the medically-relevant information. Expand
Deep Convolutional Neural Network for Inverse Problems in Imaging
• Computer Science, Mathematics
• IEEE Transactions on Image Processing
• 2017
The proposed network outperforms total variation-regularized iterative reconstruction for the more realistic phantoms and requires less than a second to reconstruct a <inline-formula> <tex-math notation="LaTeX">$512\times 512$ </tex- math></inline- formula> image on the GPU. Expand
Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization.
• Mathematics, Medicine
• Physics in medicine and biology
• 2008
An iterative algorithm, based on recent work in compressive sensing, that minimizes the total variation of the image subject to the constraint that the estimated projection data is within a specified tolerance of the available data and that the values of the volume image are non-negative is developed. Expand
Conjugate-gradient preconditioning methods for shift-variant PET image reconstruction
• Mathematics, Medicine
• IEEE Trans. Image Process.
• 1999
New preconditioners that approximate more accurately the Hessian matrices of shift-variant imaging problems are described and lead to significantly faster convergence rates for the unconstrained conjugate-gradient (CG) iteration. Expand
Principles of Computerized Tomographic Imaging
The total attenuation suffered by one beam of x-rays as it travels in a straight line through an object can be represented by a line integral. Combining a set of line integrals forms a projection.Expand