The LoDoPaB-CT Dataset: A Benchmark Dataset for Low-Dose CT Reconstruction Methods

Leuschner, Johannes; Schmidt, Maximilian; Baguer, Daniel Otero; Maaß, Peter

doi:10.48550/arxiv.1910.01113

Cited by 12 publications

(17 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparisons are provided for two low-dose CT examples: a synthetic dataset, consisting of images of random ellipses, and the LoDoPab dataset [112], which consists of human phantoms. For both datasets, CT measurements are simulated with a parallel beam geometry with a sparse-angle setup of only 30 angles and 183 projection beams, resulting in 5,490 equations and 16,384 unknowns.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Feasibility-based Fixed Point Networks

Heaton¹,

Fung²,

Gibali³

et al. 2021

Preprint

View full text Add to dashboard Cite

Inverse problems consist of recovering a signal from a collection of noisy measurements. These problems can often be cast as feasibility problems; however, additional regularization is typically necessary to ensure accurate and stable recovery with respect to data perturbations. Hand-chosen analytic regularization can yield desirable theoretical guarantees, but such approaches have limited effectiveness recovering signals due to their inability to leverage large amounts of available data. To this end, this work fuses data-driven regularization and convex feasibility in a theoretically sound manner. This is accomplished using feasibility-based fixed point networks (F-FPNs). Each F-FPN defines a collection of nonexpansive operators, each of which is the composition of a projection-based operator and a data-driven regularization operator. Fixed point iteration is used to compute fixed points of these operators, and weights of the operators are tuned so that the fixed points closely represent available data. Numerical examples demonstrate performance increases by F-FPNs when compared to standard TV-based recovery methods for CT reconstruction and a comparable neural network based on algorithm unrolling. [1]

show abstract

Section: Methodsmentioning

confidence: 99%

“…The ellipse training and test sets contain 10,000 and 1,000 pairs, respectively. We also use phantoms derived from actual human chest CT scans via the benchmark Low-Dose Parallel Beam dataset (LoDoPaB) [112]. The LoDoPab training and test sets contain 20,000 and 2,000 pairs, respectively.…”

Section: Methodsmentioning

confidence: 99%

Feasibility-based Fixed Point Networks

Heaton¹,

Fung²,

Gibali³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…al. have mentioned, the real CT data contain circular reconstructions [27] and the data must be cropped at an square inside this circle to prevent value jumps. Accordingly, we focused the middle 384×384 pixels of each slice where there were 512×512 pixels in the original one.…”

Section: Real Chest Ct Datamentioning

confidence: 99%

3D U-NetR: Low Dose Computed Tomography Reconstruction via Deep Learning and 3 Dimensional Convolutions

Gunduzalp¹,

Ünal²,

Yıldırım³

2021

Preprint

View full text Add to dashboard Cite

In this paper, we introduced a novel deep learning based reconstruction technique using the correlations of all 3 dimensions with each other by taking into account the correlation between 2-dimensional low-dose CT images. Sparse or noisy sinograms are back projected to the image domain with FBP operation, then denoising process is applied with a U-Net like 3 dimensional network called 3D U-NetR. Proposed network is trained with synthetic and real chest CT images, and 2D U-Net is also trained with the same dataset to prove the importance of the 3rd dimension. Proposed network shows better quantitative performance on SSIM and PSNR. More importantly, 3D U-NetR captures medically critical visual details that cannot be visualized by 2D network.

show abstract

“…We trained U-Net with different downsampling variants on the LoDoPaB-CT dataset [24] to perform CT reconstruction. The networks were trained on a dataset containing 35820 images and evaluated on the test set with 3553 images.…”

Section: Ct Reconstructionmentioning

confidence: 99%

Truly shift-equivariant convolutional neural networks with adaptive polyphase upsampling

Chaman¹

2021

Preprint

View full text Add to dashboard Cite

Convolutional neural networks lack shift equivariance due to the presence of downsampling layers. In image classification, adaptive polyphase downsampling (APS-D) was recently proposed to make CNNs perfectly shift invariant. However, in networks used for image reconstruction tasks, it can not by itself restore shift equivariance. We address this problem by proposing adaptive polyphase upsampling (APS-U), a nonlinear extension of conventional upsampling, which allows CNNs to exhibit perfect shift equivariance. With MRI and CT reconstruction experiments, we show that networks containing APS-D/U layers exhibit state of the art equivariance performance without sacrificing on image reconstruction quality. In addition, unlike prior methods like data augmentation and anti-aliasing, the gains in equivariance obtained from APS-D/U also extend to images outside the training distribution.

show abstract

The LoDoPaB-CT Dataset: A Benchmark Dataset for Low-Dose CT Reconstruction Methods

Cited by 12 publications

References 31 publications

Feasibility-based Fixed Point Networks

Feasibility-based Fixed Point Networks

3D U-NetR: Low Dose Computed Tomography Reconstruction via Deep Learning and 3 Dimensional Convolutions

Truly shift-equivariant convolutional neural networks with adaptive polyphase upsampling

Contact Info

Product

Resources

About