Super-resolution reconstruction in frequency, image, and wavelet domains to reduce through-plane partial voluming in MRI

Perceptual super-resolution in multiple sclerosis MRI

Introduction

Magnetic resonance imaging (MRI) is crucial for diagnosing and monitoring of multiple sclerosis (MS) as it is used to assess lesions in the brain and spinal cord. However, in real-world clinical settings, MRI scans are often acquired with thick slices, limiting their utility for automated quantitative analyses. This work presents a single-image super-resolution (SR) reconstruction framework that leverages SR convolutional neural networks (CNN) to enhance the through-plane resolution of structural MRI in people with MS (PwMS).

Methods

Our strategy involves the supervised fine-tuning of CNN architectures, guided by a content loss function that promotes perceptual quality, as well as reconstruction accuracy, to recover high-level image features.

Results

Extensive evaluation with MRI data of PwMS shows that our SR strategy leads to more accurate MRI reconstructions than competing methods. Furthermore, it improves lesion segmentation on low-resolution MRI, approaching the performance achievable with high-resolution images.

Discussion

Results demonstrate the potential of our SR framework to facilitate the use of low-resolution retrospective MRI from real-world clinical settings to investigate quantitative image-based biomarkers of MS.

Comparison of model-based versus deep learning-based image reconstruction for thin-slice T2-weighted spin-echo prostate MRI

PURPOSE

To compare a previous model-based image reconstruction (MBIR) with a newly developed deep learning (DL)-based image reconstruction for providing improved signal-to-noise ratio (SNR) in high through-plane resolution (1 mm) T2-weighted spin-echo (T2SE) prostate MRI.

METHODS

Large-area contrast and high-contrast spatial resolution of the reconstruction methods were assessed quantitatively in experimental phantom studies. The methods were next evaluated radiologically in 17 subjects at 3.0 Tesla for whom prostate MRI was clinically indicated. For each subject, the axial T2SE raw data were directed to MBIR and to the DL reconstruction at three vendor-provided levels: (L)ow, (M)edium, and (H)igh. Thin-slice images from the four reconstructions were compared using evaluation criteria related to SNR, sharpness, contrast fidelity, and reviewer preference. Results were compared using the Wilcoxon signed-rank test using Bonferroni correction, and inter-reader comparisons were done using the Cohen and Krippendorf tests.

RESULTS

Baseline contrast and resolution in phantom studies were equivalent for all four reconstruction pathways as desired. In vivo, all three DL levels (L, M, H) provided improved SNR versus MBIR. For virtually, all other evaluation criteria DL L and M were superior to MBIR. DL L and M were evaluated as superior to DL H in fidelity of contrast. For 44 of the 51 evaluations, the DL M reconstruction was preferred.

CONCLUSION

The deep learning reconstruction method provides significant SNR improvement in thin-slice (1 mm) T2SE images of the prostate while retaining image contrast. However, if taken to too high a level (DL High), both radiological sharpness and fidelity of contrast diminish.

Enhancing Knee MR Image Clarity through Image Domain Super-Resolution Reconstruction

This study introduces a hybrid analytical super-resolution (SR) pipeline aimed at enhancing the resolution of medical magnetic resonance imaging (MRI) scans. The primary objective is to overcome the limitations of clinical MRI resolution without the need for additional expensive hardware. The proposed pipeline involves three key steps: pre-processing to re-slice and register the image stacks; SR reconstruction to combine information from three orthogonal image stacks to generate a high-resolution image stack; and post-processing using an artefact reduction convolutional neural network (ARCNN) to reduce the block artefacts introduced during SR reconstruction. The workflow was validated on a dataset of six knee MRIs obtained at high resolution using various sequences. Quantitative analysis of the method revealed promising results, showing an average mean error of 1.40 ± 2.22% in voxel intensities between the SR denoised images and the original high-resolution images. Qualitatively, the method improved out-of-plane resolution while preserving in-plane image quality. The hybrid SR pipeline also displayed robustness across different MRI sequences, demonstrating potential for clinical application in orthopaedics and beyond. Although computationally intensive, this method offers a viable alternative to costly hardware upgrades and holds promise for improving diagnostic accuracy and generating more anatomically accurate models of the human body.

High-Resolution 3D MRI With Deep Generative Networks via Novel Slice-Profile Transformation Super-Resolution

High-resolution magnetic resonance imaging (MRI) sequences, such as 3D turbo or fast spin-echo (TSE/FSE) imaging, are clinically desirable but suffer from long scanning time-related blurring when reformatted into preferred orientations. Instead, multi-slice two-dimensional (2D) TSE imaging is commonly used because of its high in-plane resolution but is limited clinically by poor through-plane resolution due to elongated voxels and the inability to generate multi-planar reformations due to staircase artifacts. Therefore, multiple 2D TSE scans are acquired in various orthogonal imaging planes, increasing the overall MRI scan time. In this study, we propose a novel slice-profile transformation super-resolution (SPTSR) framework with deep generative learning for through-plane super-resolution (SR) of multi-slice 2D TSE imaging. The deep generative networks were trained by synthesized low-resolution training input via slice-profile downsampling (SP-DS), and the trained networks inferred on the slice profile convolved (SP-conv) testing input for 5.5x through-plane SR. The network output was further slice-profile deconvolved (SP-deconv) to achieve an isotropic super-resolution. Compared to SMORE SR method and the networks trained by conventional downsampling, our SPTSR framework demonstrated the best overall image quality from 50 testing cases, evaluated by two abdominal radiologists. The quantitative analysis cross-validated the expert reader study results. 3D simulation experiments confirmed the quantitative improvement of the proposed SPTSR and the effectiveness of the SP-deconv step, compared to 3D ground-truths. Ablation studies were conducted on the individual contributions of SP-DS and SP-conv, networks structure, training dataset size, and different slice profiles.

Model-based image reconstruction with wavelet sparsity regularization for through-plane resolution restoration in T2 -weighted spin-echo prostate MRI

PURPOSE

The purpose is to develop a model-based image-reconstruction method using wavelet sparsity regularization for maintaining restoration of through-plane resolution but with improved retention of SNR versus linear reconstruction using Tikhonov (TK) regularization in high through-plane resolution (1 mm) T2 -weighted spin-echo (T2SE) images of the prostate.

METHODS

A wavelet sparsity (WS)-regularized image reconstruction was developed that takes as input a set of ≈80 overlapped 3-mm-thick slices acquired using a T2SE multislice scan and typically 30 coil elements. After testing in contrast and resolution phantoms and calibration in 6 subjects, the WS reconstruction was evaluated in 16 consecutive prostate T2SE MRI exams. Results reconstructed with nominal 1-mm thickness were compared with those from the TK reconstruction with the same raw data. Results were evaluated radiologically. The ratio of magnitude of prostate signal to periprostatic muscle signal was used to assess the presence of noise reduction. Technical performance was also compared with a commercial 3D-T2SE sequence.

RESULTS

The new WS reconstruction was assessed as superior statistically to TK for overall SNR, contrast, and multiple evaluation criteria related to sharpness while retaining the high (1 mm) through-plane resolution. Wavelet sparsity tended to provide improved overall diagnostic quality versus TK, but not significantly so. In all 16 studies, the prostate-to-muscle signal ratio increased.

CONCLUSIONS

Model-based WS-regularized reconstruction consistently provides improved SNR in high (1 mm) through-plane resolution images of prostate T2SE MRI versus linear reconstruction using TK regularization.

Volumetric Fetal Flow Imaging With Magnetic Resonance Imaging

Fetal development relies on a complex circulatory network. Accurate assessment of flow distribution is important for understanding pathologies and potential therapies. In this paper, we demonstrate a method for volumetric imaging of fetal flow with magnetic resonance imaging (MRI). Fetal MRI faces challenges: small vascular structures, unpredictable motion, and inadequate traditional cardiac gating methods. Here, orthogonal multislice stacks are acquired with accelerated multidimensional radial phase contrast (PC) MRI. Slices are reconstructed into flow sensitive time-series images with motion correction and image-based cardiac gating. They are then combined into a dynamic volume using slice-to-volume reconstruction (SVR) while resolving interslice spatiotemporal coregistration. Compared to prior methods, this approach achieves higher spatiotemporal resolution ( 1×1×1 mm³, ~30 ms) with reduced scan time - important features for the quantification of flow through small fetal structures. Validation is demonstrated in adults by comparing SVR with 4D radial PCMRI (flow bias and limits of agreement: -1.1 ml/s and [-11.8 9.6] ml/s). Feasibility is demonstrated in late gestation fetuses by comparing SVR with 2D Cartesian PCMRI (flow bias and limits of agreement: -0.9 ml/min/kg and [-39.7 37.8] ml/min/kg). With SVR, we demonstrate complex flow pathways (such as parallel flow streams in the proximal inferior vena cava, preferential shunting of blood from the ductus venosus into the left atrium, and blood from the brain leaving the heart through the main pulmonary artery) for the first time in human fetal circulation. This method allows for comprehensive evaluation of the fetal circulation and enables future studies of fetal physiology.

Scan-Specific Generative Neural Network for MRI Super-Resolution Reconstruction

The interpretation and analysis of Magnetic resonance imaging (MRI) benefit from high spatial resolution. Unfortunately, direct acquisition of high spatial resolution MRI is time-consuming and costly, which increases the potential for motion artifact, and suffers from reduced signal-to-noise ratio (SNR). Super-resolution reconstruction (SRR) is one of the most widely used methods in MRI since it allows for the trade-off between high spatial resolution, high SNR, and reduced scan times. Deep learning has emerged for improved SRR as compared to conventional methods. However, current deep learning-based SRR methods require large-scale training datasets of high-resolution images, which are practically difficult to obtain at a suitable SNR. We sought to develop a methodology that allows for dataset-free deep learning-based SRR, through which to construct images with higher spatial resolution and of higher SNR than can be practically obtained by direct Fourier encoding. We developed a dataset-free learning method that leverages a generative neural network trained for each specific scan or set of scans, which in turn, allows for SRR tailored to the individual patient. With the SRR from three short duration scans, we achieved high quality brain MRI at an isotropic spatial resolution of 0.125 cubic mm with six minutes of imaging time for T2 contrast and an average increase of 7.2 dB (34.2%) in SNR to these short duration scans. Motion compensation was achieved by aligning the three short duration scans together. We assessed our technique on simulated MRI data and clinical data acquired from 15 subjects. Extensive experimental results demonstrate that our approach achieved superior results to state-of-the-art methods, while in parallel, performed at reduced cost as scans delivered with direct high-resolution acquisition.

Gradient-Guided Isotropic MRI Reconstruction from Anisotropic Acquisitions

The trade-off between image resolution, signal-to-noise ratio (SNR), and scan time in any magnetic resonance imaging (MRI) protocol is inevitable and unavoidable. Super-resolution reconstruction (SRR) has been shown effective in mitigating these factors, and thus, has become an important approach in addressing the current limitations of MRI. In this work, we developed a novel, image-based MRI SRR approach based on anisotropic acquisition schemes, which utilizes a new gradient guidance regularization method that guides the high-resolution (HR) reconstruction via a spatial gradient estimate. Further, we designed an analytical solution to propagate the spatial gradient fields from the low-resolution (LR) images to the HR image space and exploited these gradient fields over multiple scales with a dynamic update scheme for more accurate edge localization in the reconstruction. We also established a forward model of image formation and inverted it along with the proposed gradient guidance. The proposed SRR method allows subject motion between volumes and is able to incorporate various acquisition schemes where the LR images are acquired with arbitrary orientations and displacements, such as orthogonal and through-plane origin-shifted scans. We assessed our proposed approach on simulated data as well as on the data acquired on a Siemens 3T MRI scanner containing 45 MRI scans from 14 subjects. Our experimental results demonstrate that our approach achieved superior reconstructions compared to state-of-the-art methods, both in terms of local spatial smoothness and edge preservation, while, in parallel, at reduced, or at the same cost as scans delivered with direct HR acquisition.

MRI Super-Resolution Through Generative Degradation Learning

Spatial resolution plays a critically important role in MRI for the precise delineation of the imaged tissues. Unfortunately, acquisitions with high spatial resolution require increased imaging time, which increases the potential of subject motion, and suffers from reduced signal-to-noise ratio (SNR). Super-resolution reconstruction (SRR) has recently emerged as a technique that allows for a trade-off between high spatial resolution, high SNR, and short scan duration. Deconvolution-based SRR has recently received significant interest due to the convenience of using the image space. The most critical factor to succeed in deconvolution is the accuracy of the estimated blur kernels that characterize how the image was degraded in the acquisition process. Current methods use handcrafted filters, such as Gaussian filters, to approximate the blur kernels, and have achieved promising SRR results. As the image degradation is complex and varies with different sequences and scanners, handcrafted filters, unfortunately, do not necessarily ensure the success of the deconvolution. We sought to develop a technique that enables accurately estimating blur kernels from the image data itself. We designed a deep architecture that utilizes an adversarial scheme with a generative neural network against its degradation counterparts. This design allows for the SRR tailored to an individual subject, as the training requires the scan-specific data only, i.e., it does not require auxiliary datasets of high-quality images, which are practically challenging to obtain. With this technique, we achieved high-quality brain MRI at an isotropic resolution of 0.125 cubic mm with six minutes of imaging time. Extensive experiments on both simulated low-resolution data and clinical data acquired from ten pediatric patients demonstrated that our approach achieved superior SRR results as compared to state-of-the-art deconvolution-based methods, while in parallel, at substantially reduced imaging time in comparison to direct high-resolution acquisitions.

3D dense convolutional neural network for fast and accurate single MR image super-resolution

Super-resolution (SR) MR image reconstruction has shown to be a very promising direction to improve the spatial resolution of low-resolution (LR) MR images. In this paper, we presented a novel MR image SR method based on a dense convolutional neural network (DDSR), and its enhanced version called EDDSR. There are three major innovations: first, we re-designed dense modules to extract hierarchical features directly from LR images and propagate the extracted feature maps through dense connections. Therefore, unlike other CNN-based SR MR techniques that upsample LR patches in the initial phase, our methods take the original LR images or patches as input. This effectively reduces computational complexity and speeds up SR reconstruction. Second, a final deconvolution filter in our model automatically learns filters to fuse and upscale all hierarchical feature maps to generate HR MR images. Using this, EDDSR can perform SR reconstructions at different upscale factors using a single model with one stride fixed deconvolution operation. Third, to further improve SR reconstruction accuracy, we exploited a geometric self-ensemble strategy. Experimental results on three benchmark datasets demonstrate that our methods, DDSR and EDDSR, achieved superior performance compared to state-of-the-art MR image SR methods with less computational load and memory usage.

Fast and High-Resolution Neonatal Brain MRI Through Super-Resolution Reconstruction From Acquisitions With Variable Slice Selection Direction

The brain of neonates is small in comparison to adults. Imaging at typical resolutions such as one cubic mm incurs more partial voluming artifacts in a neonate than in an adult. The interpretation and analysis of MRI of the neonatal brain benefit from a reduction in partial volume averaging that can be achieved with high spatial resolution. Unfortunately, direct acquisition of high spatial resolution MRI is slow, which increases the potential for motion artifact, and suffers from reduced signal-to-noise ratio. The purpose of this study is thus that using super-resolution reconstruction in conjunction with fast imaging protocols to construct neonatal brain MRI images at a suitable signal-to-noise ratio and with higher spatial resolution than can be practically obtained by direct Fourier encoding. We achieved high quality brain MRI at a spatial resolution of isotropic 0.4 mm with 6 min of imaging time, using super-resolution reconstruction from three short duration scans with variable directions of slice selection. Motion compensation was achieved by aligning the three short duration scans together. We applied this technique to 20 newborns and assessed the quality of the images we reconstructed. Experiments show that our approach to super-resolution reconstruction achieved considerable improvement in spatial resolution and signal-to-noise ratio, while, in parallel, substantially reduced scan times, as compared to direct high-resolution acquisitions. The experimental results demonstrate that our approach allowed for fast and high-quality neonatal brain MRI for both scientific research and clinical studies.

Cross correlation-based misregistration correction for super resolution T2 -weighted spin-echo images: application to prostate

PURPOSE

The purpose is to develop a retrospective correction for subtle slice-to-slice positional inconsistencies that can occur when overlapped slices are acquired for super resolution in T2 -weighted spin-echo multislice imaging.

METHODS

Spin-echo acquisition of overlapped slices is typically done using multiple passes. After the passes are assembled into the final slice set, consecutive slices are correlated due to their overlap. Cross correlation was used to measure slice-to-slice displacement. After Z-dependent filtering to preserve true object shape, the displacements were used to correct slice position. The method was tested in a phantom moved slowly (0.16-0.63 mm/pass) under computer control and in vivo in 16 patients having prostate MRI.

RESULTS

Over the motion range, the correlation method had an accuracy within 0.03 mm/pass and precision ± 0.20 mm (ie, subpixel). Corrected images visually resemble the true object. Over the patient studies, the mean range of motion in the anterior-posterior direction was 1.63 mm. Motion-corrected axial images and the sagittal reformats were evaluated as significantly superior over those formed without motion correction.

CONCLUSION

The retrospective correlation-based motion-correction method provides significant improvement in the slice-to-slice registration necessary for effective super resolution using overlapped slices.

Modified acquisition strategy for reduced motion artifact in super resolution T 2 FSE multislice MRI: Application to prostate

PURPOSE

To reduce slice-to-slice motion effects in multislice T 2 -weighted fast-spin-echo ( T 2 FSE) imaging, manifest as "scalloping" in reformats, by modification of the acquisition strategy and to show applicability in prostate MRI.

METHODS

T 2 FSE images of contiguous or overlapping slices are typically acquired using multiple passes in which each pass is comprised of multiple slices with slice-to-slice gaps. Combination of slices from all passes provides the desired sampling. For enhancement of through-plane resolution with super resolution or for reformatting into other orientations, subtle ≈1 mm motion between passes can cause objectionable "scalloping" artifact. Here we address this by subdivision of each pass into multiple segments. Interleaving of segments from the multiple passes causes all slices to be acquired over substantially the same time, reducing pass-to-pass motion effects. This was implemented in acquiring 78 overlapped T 2 FSE axial slices and studied in phantoms and in 14 prostate MRI patients. Super-resolution axial images and sagittal reformats from the original and new segmented acquisitions were evaluated by 3 uroradiologists.

RESULTS

For all criteria of sagittal reformats, the segmented acquisition was statistically superior to the original. For all sharpness criteria of axial images, although the trend preferred the original acquisition, the difference was not significant. For artifact in axial images, the segmented acquisition was significantly superior.

CONCLUSIONS

For prostate MRI the new segmented acquisition significantly reduces the scalloping motion artifact that can be present in reformats due to long time lags between the acquisition of adjacent or overlapped slices while retaining image sharpness in the acquired axial slices.

Isotropic MRI Super-Resolution Reconstruction with Multi-scale Gradient Field Prior

In this work, we proposed a novel image-based MRI super-resolution reconstruction (SRR) approach based on anisotropic acquisition schemes. We achieved superior reconstruction to state-of-the-art work by introducing a new multi-scale gradient field prior that guides the reconstruction of the high-resolution (HR) image. The prior improves both spatial smoothness and edge preservation. The inverse of the forward model of image formation is used to propagate the gradient guidance from the low-resolution (LR) images to the HR image space. The gradient fields over multiple scales were exploited for more accurate edge localization in the reconstruction. The proposed SRR allows inter-volume motion during the MRI scans and can incorporate with the LR images with arbitrary orientations and displacements in the frequency space, such as orthogonal and origin-shifted scans. The proposed approach was evaluated on the synthetic data as well as the data acquired on a Siemens 3T MRI scanner containing 45 MRI scans from 14 subjects. The evaluation results demonstrate that our proposed prior leads to improved SRR as compared to state-of-the-art priors, and that the proposed SRR obtains better results at lower or the same cost in scan time than direct HR acquisition. In particular, the anatomical structures of hippocampus can be clearly shown in our reconstructed images. This is a significant improvement for the in vivo studies of the hippocampus.

Improved detection of erosions in the sacroiliac joints on MRI with volumetric interpolated breath-hold examination (VIBE): results from the SIMACT study

OBJECTIVE

To compare the performance of a new three-dimensional MRI sequence (volumetric interpolated breath-hold examination; MR-VIBE) with a conventional T1-weighted sequence (MR-T1) for the detection of erosions in the sacroiliac joints (SIJs) using low-dose CT (ldCT) as reference.

METHODS

ldCT and T1-MRI and MR-VIBE of 110 prospectively included patients with low back pain and suspected axial spondyloarthritis (axSpA) were scored for erosions by two readers. The presence of erosions on the patients' level, the erosion sum score, sensitivity and specificity of both MRI sequences using ldCT as a reference as well as agreement between the readers were assessed.

RESULTS

MR-VIBE had a higher sensitivity than MR-T1 (95% vs 79%, respectively) without a decrease in specificity (93% each). MR-VIBE compared with MR-T1 identified 16% more patients with erosions (36 vs 30 of 38 patients with positive ldCT findings). The erosion sum score was also higher for MR-VIBE (8.1±9.3) than MR-T1 (6.7±8.4), p=0.003. The agreement on erosion detection was also higher for MR-VIBE (κ=0.71) compared with MRI-T1 (κ=0.56).

CONCLUSION

VIBE detected erosions in the SIJs with higher sensitivity without a loss of specificity and superior reliability compared with a standard T1-weighted sequence. Its value for the diagnosis of axSpA has still to be determined.

A New Sparse Representation Framework for Reconstruction of an Isotropic High Spatial Resolution MR Volume From Orthogonal Anisotropic Resolution Scans

In magnetic resonance (MR), hardware limitations, scan time constraints, and patient movement often result in the acquisition of anisotropic 3-D MR images with limited spatial resolution in the out-of-plane views. Our goal is to construct an isotropic high-resolution (HR) 3-D MR image through upsampling and fusion of orthogonal anisotropic input scans. We propose a multiframe super-resolution (SR) reconstruction technique based on sparse representation of MR images. Our proposed algorithm exploits the correspondence between the HR slices and the low-resolution (LR) sections of the orthogonal input scans as well as the self-similarity of each input scan to train pairs of overcomplete dictionaries that are used in a sparse-land local model to upsample the input scans. The upsampled images are then combined using wavelet fusion and error backprojection to reconstruct an image. Features are learned from the data and no extra training set is needed. Qualitative and quantitative analyses were conducted to evaluate the proposed algorithm using simulated and clinical MR scans. Experimental results show that the proposed algorithm achieves promising results in terms of peak signal-to-noise ratio, structural similarity image index, intensity profiles, and visualization of small structures obscured in the LR imaging process due to partial volume effects. Our novel SR algorithm outperforms the nonlocal means (NLM) method using self-similarity, NLM method using self-similarity and image prior, self-training dictionary learning-based SR method, averaging of upsampled scans, and the wavelet fusion method. Our SR algorithm can reduce through-plane partial volume artifact by combining multiple orthogonal MR scans, and thus can potentially improve medical image analysis, research, and clinical diagnosis.

Evaluation of motion and its effect on brain magnetic resonance image quality in children

BACKGROUND

Motion artifacts pose significant problems for the acquisition of MR images in pediatric populations.

OBJECTIVE

To evaluate temporal motion metrics in MRI scanners and their effect on image quality in pediatric populations in neuroimaging studies.

MATERIALS AND METHODS

We report results from a large pediatric brain imaging study that shows the effect of motion on MRI quality. We measured motion metrics in 82 pediatric patients, mean age 13.4 years, in a T1-weighted brain MRI scan. As a result of technical difficulties, 5 scans were not included in the subsequent analyses. A radiologist graded the images using a 4-point scale ranging from clinically non-diagnostic because of motion artifacts to no motion artifacts. We used these grades to correlate motion parameters such as maximum motion, mean displacement from a reference point, and motion-free time with image quality.

RESULTS

Our results show that both motion-free time (as a ratio of total scan time) and average displacement from a position at a fixed time (when the center of k-space was acquired) were highly correlated with image quality, whereas maximum displacement was not as good a predictor. Among the 77 patients whose motion was measured successfully, 17 had average displacements of greater than 0.5 mm, and 11 of those (14.3%) resulted in non-diagnostic images. Similarly, 14 patients (18.2%) had less than 90% motion-free time, which also resulted in non-diagnostic images.

CONCLUSION

We report results from a large pediatric study to show how children and young adults move in the MRI scanner and the effect that this motion has on image quality. The results will help the motion-correction community in better understanding motion patterns in pediatric populations and how these patterns affect MR image quality.