Exploring Separable Attention for Multi-Contrast MR Image Super-Resolution

Feng, Chun-Mei; Yan, Yunlu; Liu, Chengliang; Fu, Huazhu; Xu, Yong; Shao, Ling

doi:10.48550/arxiv.2109.01664

Cited by 3 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…T 2 Net [13], and SANet [14]. The benchmark results are trained using the same ways and training datasets as described in their corresponding works.…”

Section: Resultsmentioning

confidence: 99%

“…Medical image super-resolution methods are classified into two broad categories: single-contrast super-resolution (SCSR) [13,21,22,28,30,[41][42][43] and multi-contrast super-resolution (MCSR) [14,24,27,37,44]. Traditionally, because of their simplicity, bicubic and b-spline interpolations are two of the most frequently used SCSR methods in MRI practice.…”

Section: Medical Image Super-resolutionmentioning

confidence: 99%

“…In another line of studies, MCSR methods have shown superior performance and the ability to make full use of domain knowledge from multiple modalities. For example, Feng et al [14] develop a novel and successful solution called SANet, which consists of a separable attention network for exploring foreground and background areas in forward and backward directions using auxiliary contrast.…”

Section: Medical Image Super-resolutionmentioning

confidence: 99%

“…This is particularly problematic when protocols requiring a long echo time (TE) or repetition time (TR) are used. These scans may lead to inaccurate diagnosis as limited structural and textural information is provided in subsequent quantitative medical image analysis [14]. As a result, there is emerging interest in developing super-resolution (SR) techniques for reconstructing high-resolution (HR) outputs from low-resolution (LR) images to increase the spatial resolution of magnetic resonance imaging.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Model-Guided Multi-Contrast Deep Unfolding Network for MRI Super-resolution Reconstruction

Yang

Zhang

Zhou

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Magnetic resonance imaging (MRI) with high resolution (HR) provides more detailed information for accurate diagnosis and quantitative image analysis. Despite the significant advances, most existing super-resolution (SR) reconstruction network for medical images has two flaws: 1) All of them are designed in a black-box principle, thus lacking sufficient interpretability and further limiting their practical applications. Interpretable neural network models are of significant interest since they enhance the trustworthiness required in clinical practice when dealing with medical images. 2) most existing SR reconstruction approaches only use a single contrast or use a simple multi-contrast fusion mechanism, neglecting the complex relationships between different contrasts that are critical for SR improvement. To deal with these issues, in this paper, a novel Model-Guided interpretable Deep Unfolding Network (MGDUN) for medical image SR reconstruction is proposed. The Model-Guided image SR reconstruction approach solves manually designed objective functions to reconstruct HR MRI. We show how to unfold an iterative MGDUN algorithm into a novel model-guided deep unfolding network by taking the MRI observation matrix and explicit

show abstract

“…T 2 Net [13], and SANet [14]. The benchmark results are trained using the same ways and training datasets as described in their corresponding works.…”

Section: Resultsmentioning

confidence: 99%

Section: Medical Image Super-resolutionmentioning

confidence: 99%

Section: Medical Image Super-resolutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Model-Guided Multi-Contrast Deep Unfolding Network for MRI Super-resolution Reconstruction

Yang

Zhang

Zhou

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…In the field of deep convolutional neural networks, attention mechanism plays a crucial role to reconstruct the high-quality MRI images. In this regard, Feng et al [59] suggested the idea of separable attention for Multi-contrast MRI image SR. Separable attention mechanism is used to extract the contrast anatomical information such as blood vessels, tissues, and bones. Chen et al [60] proposed trusted deep CNN based SR model known as feedback adaptive weighted dense network (FWDN) to reconstruct the high-resolution medical image from the low-resolution medical input image.…”

Section: Related Workmentioning

confidence: 99%

IRMIRS: Inception-ResNet-Based Network for MRI Image Super-Resolution

Muhammad¹,

Bhutto²,

Masroor³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Medical image super-resolution is a fundamental challenge due to absorption and scattering in tissues. These challenges are increasing the interest in the quality of medical images. Recent research has proven that the rapid progress in convolutional neural networks (CNNs) has achieved superior performance in the area of medical image super-resolution. However, the traditional CNN approaches use interpolation techniques as a preprocessing stage to enlarge low-resolution magnetic resonance (MR) images, adding extra noise in the models and more memory consumption. Furthermore, conventional deep CNN approaches used layers in series-wise connection to create the deeper mode, because this later end layer cannot receive complete information and work as a dead layer. In this paper, we propose Inception-ResNet-based Network for MRI Image Super-Resolution known as IRMRIS. In our proposed approach, a bicubic interpolation is replaced with a deconvolution layer to learn the upsampling filters. Furthermore, a residual skip connection with the Inception block is used to reconstruct a high-resolution output image from a low-quality input image. Quantitative and qualitative evaluations of the proposed method are supported through extensive experiments in reconstructing sharper and clean texture details as compared to the state-of-the-art methods.

show abstract

Diagnostic advantage of thin slice 2D MRI and multiplanar reconstruction of the knee joint using deep learning based denoising approach

Kakigi

Sakamoto

Tagawa

et al. 2022

Sci Rep

View full text Add to dashboard Cite

The purpose of this study is to evaluate whether thin-slice high-resolution 2D fat-suppressed proton density-weighted image of the knee joint using denoising approach with deep learning-based reconstruction (dDLR) with MPR is more useful than 3D FS-PD multi planar voxel image. Twelve patients who underwent MRI of the knee at 3T and 13 knees were enrolled. Denoising effect was quantitatively evaluated by comparing the coefficient of variation (CV) before and after dDLR. For the qualitative assessment, two radiologists evaluated image quality, artifacts, anatomical structures, and abnormal findings using a 5-point Likert scale between 2D and 3D. All of them were statistically analyzed. Gwet’s agreement coefficients were also calculated. For the scores of abnormal findings, we calculated the percentages of the cases with agreement with high confidence. The CV after dDLR was significantly lower than the one before dDLR (p < 0.05). As for image quality, artifacts and anatomical structure, no significant differences were found except for flow artifact (p < 0.05). The agreement was significantly higher in 2D than in 3D in abnormal findings (p < 0.05). In abnormal findings, the percentage with high confidence was higher in 2D than in 3D (p < 0.05). By applying dDLR to 2D, almost equivalent image quality to 3D could be obtained. Furthermore, abnormal findings could be depicted with greater confidence and consistency, indicating that 2D with dDLR can be a promising imaging method for the knee joint disease evaluation.

show abstract

Exploring Separable Attention for Multi-Contrast MR Image Super-Resolution

Cited by 3 publications

References 26 publications

Model-Guided Multi-Contrast Deep Unfolding Network for MRI Super-resolution Reconstruction

Model-Guided Multi-Contrast Deep Unfolding Network for MRI Super-resolution Reconstruction

IRMIRS: Inception-ResNet-Based Network for MRI Image Super-Resolution

Diagnostic advantage of thin slice 2D MRI and multiplanar reconstruction of the knee joint using deep learning based denoising approach

Contact Info

Product

Resources

About