Sinogram + image domain neural network approach for metal artifact reduction in low-dose cone-beam computed tomography

Ketcha, Michael D.; Marrama, Michael; Souza, Alex Júnior Souza de; Uneri, Ali; Wu, Pu‐Wei; Zhang, Xiaoxuan; Helm, Patrick A.; Siewerdsen, Jeffrey H.

doi:10.1117/1.jmi.8.5.052103

Cited by 17 publications

(23 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is possible because a self‐supervised target function can be formulated directly on the reconstructed volume while still driving parameter updates in the raw data domain. The cupping correction is a simple yet effective application of the proposed reconstruction pipeline which also hints at its potential in tasks requiring deeper models like complex artefact suppression 14,15 or image denoising 16,17 …”

Section: Introductionmentioning

confidence: 87%

Calibration by differentiation – Self‐supervised calibration for X‐ray microscopy using a differentiable cone‐beam reconstruction operator

Thies

Wagner

Huang

et al. 2022

Journal of Microscopy

View full text Add to dashboard Cite

High‐resolution X‐ray microscopy (XRM) is gaining interest for biological investigations of extremely small‐scale structures. XRM imaging of bones in living mice could provide new insights into the emergence and treatment of osteoporosis by observing osteocyte lacunae, which are holes in the bone of few micrometres in size. Imaging living animals at that resolution, however, is extremely challenging and requires very sophisticated data processing converting the raw XRM detector output into reconstructed images. This paper presents an open‐source, differentiable reconstruction pipeline for XRM data which analytically computes the final image from the raw measurements. In contrast to most proprietary reconstruction software, it offers the user full control over each processing step and, additionally, makes the entire pipeline deep learning compatible by ensuring differentiability. This allows fitting trainable modules both before and after the actual reconstruction step in a purely data‐driven way using the gradient‐based optimizers of common deep learning frameworks. The value of such differentiability is demonstrated by calibrating the parameters of a simple cupping correction module operating on the raw projection images using only a self‐supervisory quality metric based on the reconstructed volume and no further calibration measurements. The retrospective calibration directly improves image quality as it avoids cupping artefacts and decreases the difference in grey values between outer and inner bone by 68–94%. Furthermore, it makes the reconstruction process entirely independent of the XRM manufacturer and paves the way to explore modern deep learning reconstruction methods for arbitrary XRM and, potentially, other flat‐panel computed tomography systems. This exemplifies how differentiable reconstruction can be leveraged in the context of XRM and, hence, is an important step towards the goal of reducing the resolution limit of in vivo bone imaging to the single micrometre domain.

show abstract

Section: Introductionmentioning

confidence: 87%

Calibration by differentiation – Self‐supervised calibration for X‐ray microscopy using a differentiable cone‐beam reconstruction operator

Thies

Wagner

Huang

et al. 2022

Journal of Microscopy

View full text Add to dashboard Cite

show abstract

“…However, due to the dependency of the kernel on two voxels of the input volume X k and X i , the analytical derivative of the convolutional kernel is more elaborate. When calculating the sum in (13), both (15) and ( 16) are used for the contributions k = i and only the term k = i is derived in (17) and (18). Note that for faster computation the sums in ( 17) and ( 18) can be pre-calculated in the forward pass of the filtering.…”

Section: Conflict Of Interestmentioning

confidence: 99%

“…Whereas such approaches usually require hand-tuned hyperparameters, cannot abstract complex features, or are known to be computationally expensive, purely data-driven, end-to-end trainable, approaches have been proposed in recent years 11,12,13,14,15,16,17 fueled by the emergence of deep learning and, in particular, convolutional neural networks. Most of these models achieve competitive denoising performance but are built on deep neural networks with multiple layers and contain hundreds of thousands of trainable parameters.…”

Section: Introductionmentioning

confidence: 99%

Ultra Low-Parameter Denoising: Trainable Bilateral Filter Layers in Computed Tomography

Wagner¹,

Thies²,

Gu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Computed tomography is widely used as an imaging tool to visualize three-dimensional structures with expressive bone-soft tissue contrast. However, CT resolution and radiation dose are tightly entangled, highlighting the importance of lowdose CT combined with sophisticated denoising algorithms. Purpose: Most data-driven denoising techniques are based on deep neural networks and, therefore, contain hundreds of thousands of trainable parameters, making them incomprehensible and prone to prediction failures. Developing understandable and robust denoising algorithms achieving state-of-the-art performance helps to minimize radiation dose while maintaining data integrity. Methods: This work presents an open-source CT denoising framework based on the idea of bilateral filtering. We propose a bilateral filter that can be incorporated into a deep learning pipeline and optimized in a purely data-driven way by calculating the gradient flow toward its hyperparameters and its input. Denoising in pure image-to-image pipelines and across different domains such as raw detector data and reconstructed volume, using a differentiable backprojection layer, is demonstrated. Results:Although only using three spatial parameters and one range parameter per filter layer, the proposed denoising pipelines can compete with deep state-of-the-art denoising architectures with several hundred thousand parameters. Competitive denoising performance is achieved on x-ray microscope bone data (0.7053 and 33.10) and the 2016 Low Dose CT Grand Challenge dataset (0.9674 and 43.07) in terms of SSIM and PSNR.

show abstract

“…To handle this problem, several MAR methods have been proposed recently for sparse-view sampling schemes. [32][33][34] However, the TV-IR-based method 34 does not utilize any DL-based prior, and this shows limited performance in improving reconstructed image Compared to the above methods, our method provides state-of -the-art image quality without potential risk of the errors in sinogram-domain CNN processing. In the proposed framework, two CNNs used together with NMAR processing to facilitate artifact reduction in metal-inserted sparse-view CT images.…”

Section: Introductionmentioning

confidence: 97%

Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

et al. 2022

View full text Add to dashboard Cite

Purpose: Sparse-view sampling has attracted attention for reducing the scan time and radiation dose of dental cone-beam computed tomography (CBCT). Recently, various deep learning-based image reconstruction techniques for sparse-view CT have been employed to produce high-quality image while effectively reducing streak artifacts caused by the lack of projection views. However, most of these methods do not fully consider the effects of metal implants. As sparse-view sampling strengthens the artifacts caused by metal objects, simultaneously reducing both metal and streak artifacts in sparse-view CT images has been challenging. To solve this problem, in this study, we propose a novel framework. Methods: The proposed method was based on the normalized metal artifact reduction (NMAR) method, and its performance was enhanced using two convolutional neural networks (CNNs). The first network reduced the initial artifacts while preserving the fine details to generate high-quality priors for NMAR processing. Subsequently, the second network was employed to reduce the streak artifacts after NMAR processing of sparse-view CT data. To validate the proposed method, we generated training and test data by computer simulations using both extended cardiac-torso (XCAT) and clinical data sets. Results: Visual inspection and quantitative evaluations demonstrated that the proposed method effectively reduced both metal and streak artifacts while preserving the details of anatomical structures compared with the conventional metal artifact reduction methods. Conclusions: We propose a framework for reconstructing accurate CT images in metal-inserted sparse-view CT. The proposed method reduces streak artifacts from both metal objects and sparse-view sampling while recovering the anatomical details, indicating the feasibility of fast-scan dental CBCT imaging.

show abstract

Sinogram + image domain neural network approach for metal artifact reduction in low-dose cone-beam computed tomography

Cited by 17 publications

References 27 publications

Calibration by differentiation – Self‐supervised calibration for X‐ray microscopy using a differentiable cone‐beam reconstruction operator

Calibration by differentiation – Self‐supervised calibration for X‐ray microscopy using a differentiable cone‐beam reconstruction operator

Ultra Low-Parameter Denoising: Trainable Bilateral Filter Layers in Computed Tomography

Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

Contact Info

Product

Resources

About