Investigation of domain gap problem in several deep-learning-based CT metal artefact reduction methods

Du, Muge; Liang, Kaichao; Liu, Yinong; Xing, Yuxiang

doi:10.48550/arxiv.2111.12983

Cited by 3 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of medical images, these kind of data are hard to create. Because of that, the majority of learning-based methods are forced to use simulated data, although they oftentimes cannot completely reproduce the complex underlying physical image formation process (e.g., DeepDRR 26 ), thus ending up with a domain gap as described by Du et al 27 Alternatively, physically correct simulations require extensive computational effort in the case of, for example, Monte Carlo simulations (e.g., MC-GPU 28 ). We assume that methods, which are solely trained on simulated data, do struggle to generalize well for a variety of real clinical cases.…”

Section: Discussionmentioning

confidence: 99%

DL‐based inpainting for metal artifact reduction for cone beam CT using metal path length information

et al. 2022

View full text Add to dashboard Cite

Background Metallic implants, which are inserted into the patient's body during trauma interventions, are the main cause of heavy artifacts in 3D X‐ray acquisitions. These artifacts then hinder the evaluation of the correct implant's positioning, thus leading to a disturbed patient's healing process and increased revision rates. Purpose This problem is tackled by so‐called metal artifact reduction (MAR) methods. This paper examines possible advances in the inpainting process of such MAR methods to decrease disruptive artifacts while simultaneously preserving important anatomical structures adjacent to the inserted implants. Methods In this paper, a learning‐based inpainting method for cone‐beam computed tomography is proposed that couples a convolutional neural network (CNN) with an estimated metal path length as prior knowledge. Further, the proposed method is solely trained and evaluated on real measured data. Results The proposed inpainting approach shows advantages over the inpainting method used by the currently clinically approved frequency split metal artifact reduction (fsMAR) method as well as the learning‐based state‐of‐the‐art (SOTA) method PConv‐Net. The major improvement of the proposed inpainting method lies in the ability to correctly preserve important anatomical structures in those regions adjacent to the metal implants. Especially these regions are highly important for a correct implant's positioning in an intraoperative setup. Using the proposed inpainting, the corresponding MAR volumes reach a mean structural similarity index measure (SSIM) score of 0.9974 and outperform the other methods by up to 6 dB on single slices regarding the peak signal‐to‐noise ratio (PSNR) score. Furthermore, it can be shown that the proposed method can generalize to clinical cases at hand. Conclusions In this paper, a learning‐based inpainting network is proposed that leverages prior knowledge about the metal path length of the inserted implant. Evaluations on real measured data reveal an increased overall MAR performance, especially regarding the preservation of anatomical structures adjacent to the inserted implants. Further evaluations suggest the ability of the proposed approach to generalize to clinical cases.

show abstract

Section: Discussionmentioning

confidence: 99%

DL‐based inpainting for metal artifact reduction for cone beam CT using metal path length information

et al. 2022

View full text Add to dashboard Cite

show abstract

“…And the presented PISC method yields the optimal MAR performance in terms of more structure preservation and higher spatial resolution. Therefore, the presented PISC method can avoid the domain gap problem, which commonly exists in the DL based MAR methods (Du et al 2021). The main reason is the presented PISC method could leverage the latent structure information in raw projection data to recover the structural details near the metal implants.…”

Section: Comparison With DL Based Mar Methodsmentioning

confidence: 99%

Physics-informed sinogram completion for metal artifact reduction in CT imaging

Zhu

Song

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective. Metal artifacts in the computed tomography (CT) imaging are unavoidably adverse to the clinical diagnosis and treatment outcomes. Most metal artifact reduction (MAR) methods easily result in the over-smoothing problem and loss of structure details near the metal implants, especially for these metal implants with irregular elongated shapes. To address this problem, we present the physics-informed sinogram completion (PISC) method for MAR in CT imaging, to reduce metal artifacts and recover more structural textures. Approach. Specifically, the original uncorrected sinogram is firstly completed by the normalized linear interpolation algorithm to reduce metal artifacts. Simultaneously, the uncorrected sinogram is also corrected based on the beam-hardening correction physical model, to recover the latent structure information in metal trajectory region by leveraging the attenuation characteristics of different materials. Both corrected sinograms are fused with the pixel-wise adaptive weights, which are manually designed according to the shape and material information of metal implants. To furtherly reduce artifacts and improve the CT image quality, a post-processing frequency split algorithm is adopted to yield the final corrected CT image after reconstructing the fused sinogram. Main results. We qualitatively and quantitatively evaluated the presented PISC method on two simulated datasets and three real datasets. All results demonstrate that the presented PISC method can effectively correct the metal implants with various shapes and materials, in terms of artifact suppression and structure preservation. Significance. We proposed a sinogram-domain MAR method to compensate for the over smoothing problem existing in most MAR methods by taking advantage of the physical prior knowledge, which has the potential to improve the performance of the deep learning based MAR approaches.

show abstract

“…The main limitation with SI approach is inaccurate segmentation problem that usually appear as undersegmentation and over-segmentation of metal trace within the sinogram or projection data (18). In our previous studies, it has also been reported that the artifacts were not completely removed, or additional artifacts may be introduced due to incomplete interpolation of the projection data or sinogram (27)(28).…”

Section: Introductionmentioning

confidence: 95%

“…Recently, deep learning-based algorithm (deep MAR) has been proposed which is either applied on projection (DLP-MAR) or image (DLI-MAR) domains, or combinations of both domains known as dual-domain MAR (17)(18)(19)(20). The deep MAR methods trained the deep network using both real and simulated datasets.…”

Section: Introductionmentioning

confidence: 99%

“…The deep MAR methods trained the deep network using both real and simulated datasets. Du et al, (2021) reported that the major problem with deep MAR methods is the domain gap problem which resulted in insufficient solution for metal artifact especially on dental dataset (18). However, the interpolation method is the most widely implemented for MAR as it is much more straightforward and less computationally expensive as compared to IR technique (9,21).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Laplace-Based Interpolation Method in Reduction of Metal Artifact in Computed Tomography Imaging

Osman¹,

Sobri²,

Achuthan³

et al. 2022

MJMHS

View full text Add to dashboard Cite

Introduction: Metal artifacts can degrade the image quality of computed tomography (CT) images which lead to errors in diagnosis. This study aims to evaluate the performance of Laplace interpolation (LI) method for metal artifacts reduction (MAR) in CT images in comparison with cubic spline (CS) interpolation. Methods: In this study, the proposed MAR algorithm was developed using MATLAB platform. Firstly, the virtual sinogram was acquired from CT image using Radon transform function. Then, dual-adaptive thresholding detected and segmented the metal part within the CT sinogram. Performance of the two interpolation methods to replace the missing part of segmented sinogram were evaluated. The interpolated sinogram was reconstructed, prior to image fusion to obtain the final corrected image. The qualitative and quantitative evaluations were performed on the corrected CT images (both phantom and clinical images) to evaluate the effectiveness of the proposed MAR technique. Results: From the findings, LI method had successfully replaced the missing data on both simple and complex thresholded sinogram as compared to CS method (p-value = 0.17). The artifact index was significantly reduced by LI method (p-value = 0.02). For qualitative analysis, the mean scores by radiologists for LI-corrected images were higher than original image and CS-corrected images. Conclusion: In conclusion, LI method for MAR produced better results as compared to CS interpolation method, as it worked more effective by successfully interpolated all the missing data within sinogram in most of the CT images.

show abstract

Investigation of domain gap problem in several deep-learning-based CT metal artefact reduction methods

Cited by 3 publications

References 31 publications

DL‐based inpainting for metal artifact reduction for cone beam CT using metal path length information

DL‐based inpainting for metal artifact reduction for cone beam CT using metal path length information

Physics-informed sinogram completion for metal artifact reduction in CT imaging

Laplace-Based Interpolation Method in Reduction of Metal Artifact in Computed Tomography Imaging

Contact Info

Product

Resources

About