Automated FDK-Filter Selection for Cone-Beam CT in Research Environments

Lagerwerf, Marinus J.; Palenstijn, Willem Jan; Kohr, Holger; Batenburg, Kees Joost

doi:10.1109/tci.2020.2971136

Cited by 7 publications

(9 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [ 22 , 23 , 33 ], exponential binning is used to approximate filters, leading to

coefficients to describe a filter. This approximation can be seen as a matrix

applied to a coefficient vector

…”

Section: Methodsmentioning

confidence: 99%

“…While iterative methods have been shown to be more accurate for noisy and limited data problems [ 13 , 14 , 15 , 16 , 17 , 18 ], they have a significantly higher computational cost. Consequently, there have been efforts to improve the accuracy of direct methods by computing data-specific or scanner-specific filters [ 19 , 20 , 21 , 22 , 23 ]. Although these strategies do improve the reconstruction accuracy, they also add significant computational effort or are specific to one modality, e.g., tomosynthesis [ 24 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Computationally Efficient Reconstruction Algorithm for Circular Cone-Beam Computed Tomography Using Shallow Neural Networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

Circular cone-beam (CCB) Computed Tomography (CT) has become an integral part of industrial quality control, materials science and medical imaging. The need to acquire and process each scan in a short time naturally leads to trade-offs between speed and reconstruction quality, creating a need for fast reconstruction algorithms capable of creating accurate reconstructions from limited data. In this paper, we introduce the Neural Network Feldkamp–Davis–Kress (NN-FDK) algorithm. This algorithm adds a machine learning component to the FDK algorithm to improve its reconstruction accuracy while maintaining its computational efficiency. Moreover, the NN-FDK algorithm is designed such that it has low training data requirements and is fast to train. This ensures that the proposed algorithm can be used to improve image quality in high-throughput CT scanning settings, where FDK is currently used to keep pace with the acquisition speed using readily available computational resources. We compare the NN-FDK algorithm to two standard CT reconstruction algorithms and to two popular deep neural networks trained to remove reconstruction artifacts from the 2D slices of an FDK reconstruction. We show that the NN-FDK reconstruction algorithm is substantially faster in computing a reconstruction than all the tested alternative methods except for the standard FDK algorithm and we show it can compute accurate CCB CT reconstructions in cases of high noise, a low number of projection angles or large cone angles. Moreover, we show that the training time of an NN-FDK network is orders of magnitude lower than the considered deep neural networks, with only a slight reduction in reconstruction accuracy.

show abstract

“…In [ 22 , 23 , 33 ], exponential binning is used to approximate filters, leading to

coefficients to describe a filter. This approximation can be seen as a matrix

applied to a coefficient vector

…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Computationally Efficient Reconstruction Algorithm for Circular Cone-Beam Computed Tomography Using Shallow Neural Networks

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our implementation-adapted filters are thus minimumresidual filters that have been optimized to each implementation I. The main difference between the previous works (Pelt & Batenburg, 2014;Lagerwerf et al, 2020a) and our present study is that we use a fixed forward operator in our optimization problem, which is not necessarily the transpose of the backprojection operator. More importantly, our goal in this paper is not the improvement of reconstruction accuracy, but the reduction of differences in reconstruction between various software implementations.…”

Section: Research Papers 3 Implementation-adapted Filtersmentioning

confidence: 99%

“…pÞ: Out of all reconstructions that an implemented algorithm can produce for a given dataset p by varying the filter, this reconstruction, r Ã I , is the one that results in the smallest residual error. Such filters are known as minimum-residual filters and have previously been proposed to improve reconstructions of real-space analytical algorithms in low-dose settings (Pelt & Batenburg, 2014;Lagerwerf et al, 2020a).…”

Section: Research Papers 3 Implementation-adapted Filtersmentioning

confidence: 99%

Improving reproducibility in synchrotron tomography using implementation-adapted filters

Ganguly¹,

Pelt²,

Gürsoy³

et al. 2021

J Synchrotron Radiat

View full text Add to dashboard Cite

For reconstructing large tomographic datasets fast, filtered backprojection-type or Fourier-based algorithms are still the method of choice, as they have been for decades. These robust and computationally efficient algorithms have been integrated in a broad range of software packages. The continuous mathematical formulas used for image reconstruction in such algorithms are unambiguous. However, variations in discretization and interpolation result in quantitative differences between reconstructed images, and corresponding segmentations, obtained from different software. This hinders reproducibility of experimental results, making it difficult to ensure that results and conclusions from experiments can be reproduced at different facilities or using different software. In this paper, a way to reduce such differences by optimizing the filter used in analytical algorithms is proposed. These filters can be computed using a wrapper routine around a black-box implementation of a reconstruction algorithm, and lead to quantitatively similar reconstructions. Use cases for this approach are demonstrated by computing implementation-adapted filters for several open-source implementations and applying them to simulated phantoms and real-world data acquired at the synchrotron. Our contribution to a reproducible reconstruction step forms a building block towards a fully reproducible synchrotron tomography data processing pipeline.

show abstract

“…Since the filter width is proportional to the number of pixels in each detector row, we have N e = O(log N f ) = O(log N ). This technique yields suitable filter approximations, as observed in [15,12]. Second, training may be accelerated by sampling a subset of voxels on which to minimize Equation (7), rather than the full volume.…”

Section: Algorithm 1 Nn-fbp Reconstruction Algorithmmentioning

confidence: 99%

Noise2Filter: fast, self-supervised learning and real-time reconstruction for 3D Computed Tomography

Lagerwerf¹,

Hendriksen²,

Buurlage³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

At X-ray beamlines of synchrotron light sources, the achievable timeresolution for 3D tomographic imaging of the interior of an object has been reduced to a fraction of a second, enabling rapidly changing structures to be examined. The associated data acquisition rates require sizable computational resources for reconstruction. Therefore, full 3D reconstruction of the object is usually performed after the scan has completed. Quasi-3D reconstruction -where several interactive 2D slices are computed instead of a 3D volume -has been shown to be significantly more efficient, and can enable the real-time reconstruction and visualization of the interior. However, quasi-3D reconstruction relies on filtered backprojection type algorithms, which are typically sensitive to measurement noise. To overcome this issue, we propose Noise2Filter, a learned filter method that can be trained using only the measured data, and does not require any additional training data. This method combines quasi-3D reconstruction, learned filters, and self-supervised learning to derive a tomographic reconstruction method that can be trained in under a minute and evaluated in real-time. We show limited loss of accuracy compared to training with additional training data, and improved accuracy compared to standard filter-based methods.

show abstract

Automated FDK-Filter Selection for Cone-Beam CT in Research Environments

Cited by 7 publications

References 19 publications

A Computationally Efficient Reconstruction Algorithm for Circular Cone-Beam Computed Tomography Using Shallow Neural Networks

A Computationally Efficient Reconstruction Algorithm for Circular Cone-Beam Computed Tomography Using Shallow Neural Networks

Improving reproducibility in synchrotron tomography using implementation-adapted filters

Noise2Filter: fast, self-supervised learning and real-time reconstruction for 3D Computed Tomography

Contact Info

Product

Resources

About