Sinogram interpolation for sparse-view micro-CT with deep learning neural network

  title={Sinogram interpolation for sparse-view micro-CT with deep learning neural network},
  author={Xu Dong and Swapnil Vekhande and Guohua Cao},
  booktitle={Medical Imaging},
In sparse-view Computed Tomography (CT), only a small number of projection images are taken around the object, and sinogram interpolation method has a significant impact on final image quality. When the amount of sparsity - the amount of missing views in sinogram data – is not high, conventional interpolation methods have yielded good results. When the amount of sparsity is high, more advanced sinogram interpolation methods are needed. Recently, several deep learning (DL) based sinogram… 

Artifact Reduction for Sparse-View CT Using Deep Learning With Band Patch

A lightweight fully convolutional network is developed to estimate a fully sampled sinogram from a sparse-view sinogram by enlargement in the vertical direction and a band patch is introduced as an input image for the network based on the sinogram’s characteristics.

LU-Net: combining LSTM and U-Net for sinogram synthesis in sparse-view SPECT reconstruction

A network architecture combining Long Short-Term Memory network (LSTM) and U-Net is proposed, dubbed LU-Net, to learn the mapping from sparse-view projection data to full-view data, and can learn the sequence feature of projection data at adjacent angles to synthesize the missing views in the sinogram.

Artifact Removal in Sparse-Angle CT Based on Feature Fusion Residual Network

  • Shipeng XieTao Yang
  • Computer Science
    IEEE Transactions on Radiation and Plasma Medical Sciences
  • 2021
A feature fusion residual network (FFRN), which achieves excellent performance in removing artifacts from different anatomical regions of sparse angle CT images, and is verified that artifacts in the sparse-angle CT images can be better predicted without changing the image size.

Sparse-view CT reconstruction based on fusion learning in hybrid domain

  • H. TianLing Li Fazhi Qi
  • Computer Science
    Proceedings of International Symposium on Grids & Clouds 2022 — PoS(ISGC2022)
  • 2022
Experimental results indicate that the proposed hybrid domain method based on fusion learning is practical and effective for reducing the artifacts and preserving the quality of the reconstructed image.

The use of deep learning methods in low-dose computed tomography image reconstruction: a systematic review

Algorithms using deep learning technology are superior to traditional IR methods in noise suppression, artifact reduction and structure preservation, in terms of improving the image quality of low-dose reconstructed images.

RAD-UNet: a Residual, Attention-Based, Dense UNet for CT Sparse Reconstruction

The experimental results show that the RAD-UNet can improve the reconstruction accuracy compared with three existing representative deep networks and may not only suppress streak artifacts but also better preserve image details.

Integrating Data and Image Domain Deep Learning for Limited Angle Tomography using Consensus Equilibrium

Conditional generative adversarial networks (cGANs) are used in both the data and the image domain for enhanced performance and efficient computation and combine them through a consensus process in a two-step process based on the consensus equilibrium (CE) framework.

Deep Learning Single View Computed Tomography Guided by FBP Algorithm

  • Jianqiao YuHui LiangYi Sun
  • Computer Science, Physics
    2022 12th International Conference on Information Science and Technology (ICIST)
  • 2022
This paper proposes a deep learning approach for single-view CT reconstruction guided by Filtered Back Projection (FBP) algorithm which makes the single view reconstruction accurate, fast and interpretable.

2-Step Sparse-View CT Reconstruction with a Domain-Specific Perceptual Network

A novel framework for sparse-view tomography is presented by decoupling the reconstruction into two steps using a super-resolution network, SIN, trained on the sparse projections, and a light-weight variant of the perceptual-loss that enhances domain-specific information, boosting restoration accuracy.

A Lightweight Dual-Domain Attention Framework for Sparse-View CT Reconstruction

Experiments indicate that the CAGAN strikes an excellent balance between model complexity and performance, and the pipeline outperforms the DD-Net and the DuDoNet.



Sinogram synthesis using convolutional-neural-network for sparsely view-sampled CT

A previously developed deep-learning based sinogram synthesis method is improved by adding new layers and modifying the size of receptive field in the deep learning network to overcome the above limitations and shows that the new approach synthesizes more accurate sinogram and thus leads to higher quality of CT image than the previous one.

View-interpolation of sparsely sampled sinogram using convolutional neural network

An interpolation method using convolutional neural network (CNN), which is one of the widely used deep-learning methods, to find missing projection data is developed and compared its performances with the other interpolation techniques.

A Sparse-View CT Reconstruction Method Based on Combination of DenseNet and Deconvolution

The proposed DD-Net method for sparse-view CT reconstruction achieved a competitive performance relative to the state-of-the-art methods in terms of streaking artifacts removal and structure preservation and can increase the structure similarity by up to 18% and reduce the root mean square error by up-to-42%.

Improve angular resolution for sparse-view CT with residual convolutional neural network

This work proposed a method of angular resolution recovery in projection domain based on deep residual convolutional neural network (CNN) so that projections at unmeasured views can be estimated accurately and reconstructed images have little streaking artifacts.

Deep Learning-Based Sinogram Completion for Low-Dose CT

This work examines the potential of a deep-learning-based method to construct a mapping from the observed, sparsely sampled, CT projection data to a set of densely sampled projection estimates directly in the original projection domain and casts the problem as one of sinogram in-painting and focuses on completing the projection data prior to reconstructing the images.

SIPID: A deep learning framework for sinogram interpolation and image denoising in low-dose CT reconstruction

This work proposes a hybrid deep learning approach that combines sinogram interpolation with image denoising, referred to as SIPID, and highlights that the design of sinograms interpolation network can be a promising component in CT reconstruction, since it can also seamlessly fit to all kinds of image denoing networks.

Deep Learning Computed Tomography: Learning Projection-Domain Weights From Image Domain in Limited Angle Problems

A new type of cone-beam back-projection layer is proposed, efficiently calculating the forward pass of this layer, and it is shown that the learned algorithm can be interpreted using known concepts from cone beam reconstruction: the network is able to automatically learn strategies such as compensation weights and apodization windows.

Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysis

A novel deep residual learning approach for sparse view CT reconstruction using sparse projection views based on a novel persistent homology analysis showing that the manifold of streaking artifacts is topologically simpler than original ones is developed.

Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion

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.

Effects of sparse sampling schemes on image quality in low-dose CT.

In CS-based image reconstructions both sampling density and data incoherence affect the image quality, and the authors suggest that a sampling scheme should be devised and optimized by use of these indicators.