IMPULSED model based cytological feature estimation with U-Net: Application to human brain tumor at 3T.

PURPOSE: This work introduces and validates a deep-learning-based fitting method, which can rapidly provide accurate and robust estimation of cytological features of brain tumor based on the IMPULSED (imaging microstructural parameters using limited spectrally edited diffusion) model fitting with diffusion-weighted MRI data. METHODS: The U-Net was applied to rapidly quantify extracellular diffusion coefficient (Dex ), cell size (d), and intracellular volume fraction (vin ) of brain tumor. At the training stage, the image-based training data, synthesized by randomizing quantifiable microstructural parameters within specific ranges, was used to train U-Net. At the test stage, the pre-trained U-Net was applied to estimate the microstructural parameters from simulated data and the in vivo data acquired on patients at 3T. The U-Net was compared with conventional non-linear least-squares (NLLS) fitting in simulations in terms of estimation accuracy and precision. RESULTS: Our results confirm that the proposed method yields better fidelity in simulations and is more robust to noise than the NLLS fitting. For in vivo data, the U-Net yields obvious quality improvement in parameter maps, and the estimations of all parameters are in good agreement with the NLLS fitting. Moreover, our method is several orders of magnitude faster than the NLLS fitting (from about 5 min to <1 s). CONCLUSION: The image-based training scheme proposed herein helps to improve the quality of the estimated parameters. Our deep-learning-based fitting method can estimate the cell microstructural parameters fast and accurately.

Radial magnetic resonance image reconstruction with a deep unrolled projected fast iterative soft-thresholding network.

Radially sampling of magnetic resonance imaging (MRI) is an effective way to accelerate the imaging. How to preserve the image details in reconstruction is always challenging. In this work, a deep unrolled neural network is designed to emulate the iterative sparse image reconstruction process of a projected fast soft-threshold algorithm (pFISTA). The proposed method, an unrolled pFISTA network for Deep Radial MRI (pFISTA-DR), include the preprocessing module to refine coil sensitivity maps and initial reconstructed image, the learnable convolution filters to extract image feature maps, and adaptive threshold to robustly remove image artifacts. Experimental results show that, among the compared methods, pFISTA-DR provides the best reconstruction and achieved the highest PSNR, the highest SSIM and the lowest reconstruction errors.

Spiral time-of-flight magnetic resonance angiography for intracranial vascular imaging: performance compared to conventional Cartesian angiogram.

Background: Computed tomography angiography (CTA) and digital subtraction angiography (DSA) usually raise the risk of potential malignancies with cumulative radiation doses. Current time-of-flight magnetic resonance angiography (TOF-MRA) (dubbed as cTOF), which is based on Cartesian sampling mode, may show limited diagnostic conspicuity at sinuous or branching regions. It is also prone to relatively high false positive diagnoses and undesirable display of distal intracranial vessels. This study aimed to use spiral TOF-MRA (sTOF) as a noninvasive alternative to explore possible improvement, such that the application of magnetic resonance angiography (MRA) can be extended to facilitate clinical examination or cerebrovascular disease diagnosis and follow-up studies. Methods: Initially, 37 patients with symptoms of dizziness or transient ischemic attack were consecutively recruited for suspected intracranial vascular disease examination from Zhongshan Hospital of Xiamen University between July 2020 and April 2021 in this cross-sectional prospective study. After excluding 1 patient with severe scanning artifacts, 1 patient whose scanning scope did not meet the requirement, and 1 patient with confounding tumor lesions, a total of 34 participants were included according to the inclusion and exclusion criteria. Each participant underwent intracranial vascular imaging with both sTOF and cTOF sequences on a 3.0 T MR scanner with a conventional head-neck coil of 16 channels. Contrast CTA or DSA was also performed for 15 patients showing pathology. Qualitative comparisons in terms of image quality and diagnostic efficacy ratings, quantitative comparisons in terms of signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), vessel length, and sharpness were evaluated. Pair-wise Wilcoxon test was performed to evaluate the imaging quality derived from cTOF and sTOF acquisitions and weighted Cohen's Kappa was conducted to assess the rating consistency between different physicians. Results: Compared to cTOF, sTOF showed better performance with fewer artifacts. It can effectively alleviate false positives of normal vessels being misdiagnosed as aneurysm or stenosis. Improved conspicuity was observed in cerebral distal regions with more clearly identifiable vasculature at finer scales. Quantitative comparisons in selected regions revealed significant improvement of sTOF in SNR (P<0.01 or P<0.001), CNR (P<0.001), vessel length (P<0.001), and sharpness (P<0.001) as compared to cTOF. Besides, sTOF can depict details of M1 and M2 segments of middle cerebral artery (MCA) at metallic implant region, showing its resistance to magnetic susceptibility. Conclusions: The sTOF shows higher imaging quality and lesion detectability with reduced artifacts and false positives, representing a potentially feasible surrogate in intracranial vascular imaging for future clinic routines.

FlexDTI: flexible diffusion gradient encoding scheme-based highly efficient diffusion tensor imaging using deep learning.

Objective. Most deep neural network-based diffusion tensor imaging methods require the diffusion gradients' number and directions in the data to be reconstructed to match those in the training data. This work aims to develop and evaluate a novel dynamic-convolution-based method called FlexDTI for highly efficient diffusion tensor reconstruction with flexible diffusion encoding gradient scheme.Approach. FlexDTI was developed to achieve high-quality DTI parametric mapping with flexible number and directions of diffusion encoding gradients. The method used dynamic convolution kernels to embed diffusion gradient direction information into feature maps of the corresponding diffusion signal. Furthermore, it realized the generalization of a flexible number of diffusion gradient directions by setting the maximum number of input channels of the network. The network was trained and tested using datasets from the Human Connectome Project and local hospitals. Results from FlexDTI and other advanced tensor parameter estimation methods were compared.Main results. Compared to other methods, FlexDTI successfully achieves high-quality diffusion tensor-derived parameters even if the number and directions of diffusion encoding gradients change. It reduces normalized root mean squared error by about 50% on fractional anisotropy and 15% on mean diffusivity, compared with the state-of-the-art deep learning method with flexible diffusion encoding gradient scheme.Significance. FlexDTI can well learn diffusion gradient direction information to achieve generalized DTI reconstruction with flexible diffusion gradient scheme. Both flexibility and reconstruction quality can be taken into account in this network.

CloudBrain-MRS: An intelligent cloud computing platform for in vivo magnetic resonance spectroscopy preprocessing, quantification, and analysis.

Magnetic resonance spectroscopy (MRS) is an important clinical imaging method for diagnosis of diseases. MRS spectrum is used to observe the signal intensity of metabolites or further infer their concentrations. Although the magnetic resonance vendors commonly provide basic functions of spectrum plots and metabolite quantification, the spread of clinical research of MRS is still limited due to the lack of easy-to-use processing software or platform. To address this issue, we have developed CloudBrain-MRS, a cloud-based online platform that provides powerful hardware and advanced algorithms. The platform can be accessed simply through a web browser, without the need of any program installation on the user side. CloudBrain-MRS also integrates the classic LCModel and advanced artificial intelligence algorithms and supports batch preprocessing, quantification, and analysis of MRS data from different vendors. Additionally, the platform offers useful functions: (1) Automatically statistical analysis to find biomarkers for diseases; (2) Consistency verification between the classic and artificial intelligence quantification algorithms; (3) Colorful three-dimensional visualization for easy observation of individual metabolite spectrum. Last, data of both healthy subjects and patients with mild cognitive impairment are used to demonstrate the functions of the platform. To the best of our knowledge, this is the first cloud computing platform for in vivo MRS with artificial intelligence processing. We have shared our cloud platform at MRSHub, providing at least two years of free access and service. If you are interested, please visit https://mrshub.org/software_all/#CloudBrain-MRS or https://csrc.xmu.edu.cn/CloudBrain.html.

Magnetic Resonance Spectroscopy Quantification Aided by Deep Estimations of Imperfection Factors and Macromolecular Signal.

OBJECTIVE: Magnetic Resonance Spectroscopy (MRS) is an important technique for biomedical detection. However, it is challenging to accurately quantify metabolites with proton MRS due to serious overlaps of metabolite signals, imperfections because of non-ideal acquisition conditions, and interference with strong background signals mainly from macromolecules. The most popular method, LCModel, adopts complicated non-linear least square to quantify metabolites and addresses these problems by designing empirical priors such as basis-sets, imperfection factors. However, when the signal-to-noise ratio of MRS signal is low, the solution may have large deviation. METHODS: Linear Least Squares (LLS) is integrated with deep learning to reduce the complexity of solving this overall quantification. First, a neural network is designed to explicitly predict the imperfection factors and the overall signal from macromolecules. Then, metabolite quantification is solved analytically with the introduced LLS. In our Quantification Network (QNet), LLS takes part in the backpropagation of network training, which allows the feedback of the quantification error into metabolite spectrum estimation. This scheme greatly improves the generalization to metabolite concentrations unseen in training compared to the end-to-end deep learning method. RESULTS: Experiments show that compared with LCModel, the proposed QNet, has smaller quantification errors for simulated data, and presents more stable quantification for 20 healthy in vivo data at a wide range of signal-to-noise ratio. QNet also outperforms other end-to-end deep learning methods. CONCLUSION: This study provides an intelligent, reliable and robust MRS quantification. SIGNIFICANCE: QNet is the first LLS quantification aided by deep learning.

Intravoxel incoherent motion magnetic resonance imaging reconstruction from highly under-sampled diffusion-weighted PROPELLER acquisition data via physics-informed residual feedback unrolled network.

Objective. The acquisition of diffusion-weighted images for intravoxel incoherent motion (IVIM) imaging is time consuming. This work aims to accelerate the scan through a highly under-sampling diffusion-weighted turbo spin echo PROPELLER (DW-TSE-PROPELLER) scheme and to develop a reconstruction method for accurate IVIM parameter mapping from the under-sampled data.Approach.The proposed under-sampling DW-TSE-PROPELLER scheme for IVIM imaging is that a few blades perb-value are acquired and rotated along theb-value dimension to cover high-frequency information. A physics-informed residual feedback unrolled network (PIRFU-Net) is proposed to directly estimate distortion-free and artifact-free IVIM parametric maps (i.e., the perfusion-free diffusion coefficientDand the perfusion fractionf) from highly under-sampled DW-TSE-PROPELLER data. PIRFU-Net used an unrolled convolution network to explore data redundancy in the k-q space to remove under-sampling artifacts. An empirical IVIM physical constraint was incorporated into the network to ensure that the signal evolution curves along theb-value follow a bi-exponential decay. The residual between the realistic and estimated measurements was fed into the network to refine the parametric maps. Meanwhile, the use of synthetic training data eliminated the need for genuine DW-TSE-PROPELLER data.Main results.The experimental results show that the DW-TSE-PROPELLER acquisition was six times faster than full k-space coverage PROPELLER acquisition and within a clinically acceptable time. Compared with the state-of-the-art methods, the distortion-freeDandfmaps estimated by PIRFU-Net were more accurate and had better-preserved tissue boundaries on a simulated human brain and realistic phantom/rat brain/human brain data.Significance.Our proposed method greatly accelerates IVIM imaging. It is capable of directly and simultaneously reconstructing distortion-free, artifact-free, and accurateDandfmaps from six-fold under-sampled DW-TSE-PROPELLER data.

Learning from synthetic data for reference-free Nyquist ghost correction and parallel imaging reconstruction of echo planar imaging.

BACKGROUND: Echo planar imaging (EPI) suffers from Nyquist ghost caused by eddy currents and other non-ideal factors. Deep learning has received interest for EPI ghost correction. However, large datasets with qualified labels are usually unavailable, especially for the under-sampled EPI data due to the imperfection of traditional ghost correction algorithms. PURPOSE: To develop a multi-coil synthetic-data-based deep learning method for the Nyquist ghost correction and reconstruction of under-sampled EPI. METHODS: Our network is trained purely with synthetic data. The labels of the training samples are generated by combining a public magnetic resonance imaging dataset and a few pre-collected coil sensitivity maps. The input is synthesized by under-sampling (for the accelerated case) and adding phase errors between the even and odd echoes of the label. To bridge the gap between synthetic data and data from real acquisition, linear and non-linear 2D phase errors are considered during the training data generation. RESULTS: The proposed method outperformed the existing mainstream approaches in several experiments. The average ghost-to-signal ratios of our/3-line navigator-based methods were 0.51%/5.36% and 0.42%/8.64% in fully-sampled and under-sampled in vivo experiments, respectively. In the sagittal experiments, our method successfully corrected higher-order and 2D phase errors. Our method also outperformed other reference-based methods on motion-corrupted data. In the simulation experiments, the peak signal-to-noise ratios were 37.6/38.3 dB for 2D linear/non-linear simulated phase errors, indicating that our method was consistently reliable for different kinds of phase errors. CONCLUSION: Our method achieves superb ghost correction and parallel imaging reconstruction without any calibration information, and can be readily adapted to other EPI-based applications.

A convergence analysis for projected fast iterative soft-thresholding algorithm under radial sampling MRI.

Radial sampling is a fast magnetic resonance imaging technique. Further imaging acceleration can be achieved with undersampling but how to reconstruct a clear image with fast algorithm is still challenging. Previous work has shown the advantage of removing undersampling image artifacts using the tight-frame sparse reconstruction model. This model was further solved with a projected fast iterative soft-thresholding algorithm (pFISTA). However, the convergence of this algorithm under radial sampling has not been clearly set up. In this work, the authors derived a theoretical convergence condition for this algorithm. This condition was approximated by estimating the maximal eigenvalue of reconstruction operators through the power iteration. Based on the condition, an optimal step size was further suggested to allow the fastest convergence. Verifications were made on the prospective in vivo data of static brain imaging and dynamic contrast-enhanced liver imaging, demonstrating that the recommended parameter allowed fast convergence in radial MRI.

Correlation between lipid-lowering therapy and cerebral microbleeds.

OBJECTIVE: To investigate if there is a correlation between lipid-lowering treatment with statins and the occurrence, number, and location of cerebral microbleeds (CMBs) among patients with ischemic cerebrovascular disease (ICVD), and also to compare treatment with atorvastatin and rosuvastatin in terms of the occurrence of CMBs and their differences. METHODS: In this retrospective study, we included patients who were diagnosed with ICVD and underwent susceptibility weighted imaging (SWI) in a grade A tertiary hospital from October 1, 2014 to October 1, 2022. We collected information on previous statin use, past medical history, clinical test indicators, and imaging data. RESULTS: We found that out of 522 patients, 310 patients (59.4%) had no CMB and 212 patients (40.6%) had CMBs. There was no statistically significant correlation between prior statin use, the occurrence, and number of CMBs in patients diagnosed with ICVD (P < 0.05). As for the location of CMB, there was a statistically significant correlation between prior statin use and lobar CMBs (P < 0.048). However, there was no statistically significant correlation between the use of atorvastatin and rosuvastatin and the occurrence of CMBs (P > 0.05). CONCLUSION: There was no independent correlation between previous statin use, and the occurrence, and number of CMBs in patients with ICVD. As for CMBs in different locations, there was a correlation between previous use of statin and lobar CMBs. There was no significant difference between atorvastatin and rosuvastatin in the occurrence of CMBs in patients with ICVD.

Sub-second whole brain T2mapping via multiband SENSE multiple overlapping-echo detachment imaging and deep learning.

Objective. Most quantitative magnetic resonance imaging (qMRI) methods are time-consuming. Multiple overlapping-echo detachment (MOLED) imaging can achieve quantitative parametric mapping of a single slice within around one hundred milliseconds. Nevertheless, imaging the whole brain, which involves multiple slices, still takes a few seconds. To further accelerate qMRI, we introduce multiband SENSE (MB-SENSE) technology to MOLED to realize simultaneous multi-slice T2mapping.Approach.The multiband MOLED (MB-MOLED) pulse sequence was carried out to acquire raw overlapping-echo signals, and deep learning was utilized to reconstruct T2maps. To address the issue of image quality degradation due to a high multiband factor MB, a plug-and-play (PnP) algorithm with prior denoisers (DRUNet) was applied. U-Net was used for T2map reconstruction. Numerical simulations, water phantom experiments and human brain experiments were conducted to validate our proposed approach.Main results.Numerical simulations show that PnP algorithm effectively improved the quality of reconstructed T2maps at low signal-to-noise ratios. Water phantom experiments indicate that MB-MOLED inherited the advantages of MOLED and its results were in good agreement with the results of reference method.In vivoexperiments for MB = 1, 2, 4 without the PnP algorithm, and 4 with PnP algorithm indicate that the use of PnP algorithm improved the quality of reconstructed T2maps at a high MB. For the first time, with MB = 4, T2mapping of the whole brain was achieved within 600 ms.Significance.MOLED and MB-SENSE can be combined effectively. This method enables sub-second T2mapping of the whole brain. The PnP algorithm can improve the quality of reconstructed T2maps. The novel approach shows significant promise in applications necessitating high temporal resolution, such as functional and dynamic qMRI.

A Paired Phase and Magnitude Reconstruction for Advanced Diffusion-Weighted Imaging.

OBJECTIVE: Multi-shot interleaved echo planer imaging (Ms-iEPI) can obtain diffusion-weighted images (DWI) with high spatial resolution and low distortion, but suffers from ghost artifacts introduced by phase variations between shots. In this work, we aim at solving the ms-iEPI DWI reconstructions under inter-shot motions and ultra-high b-values. METHODS: An iteratively joint estimation model with paired phase and magnitude priors is proposed to regularize the reconstruction (PAIR). The former prior is low-rankness in the k-space domain. The latter explores similar edges among multi-b-value and multi-direction DWI with weighted total variation in the image domain. The weighted total variation transfers edge information from the high SNR images (b-value = 0) to DWI reconstructions, achieving simultaneously noise suppression and image edges preservation. RESULTS: Results on simulated and in vivo data show that PAIR can remove inter-shot motion artifacts very well (8 shots) and suppress the noise under the ultra-high b-value (4000 s/mm2) significantly. CONCLUSION: The joint estimation model PAIR with complementary priors has a good performance on challenging reconstructions under inter-shot motions and a low signal-to-noise ratio. SIGNIFICANCE: PAIR has potential in advanced clinical DWI applications and microstructure research.

Resting-State Neuronal Activity and Functional Connectivity Changes in the Visual Cortex after High Altitude Exposure: A Longitudinal Study.

Damage to the visual cortex structures after high altitude exposure has been well clarified. However, changes in the neuronal activity and functional connectivity (FC) of the visual cortex after hypoxia/reoxygenation remain unclear. Twenty-three sea-level college students, who took part in 30 days of teaching at high altitude (4300 m), underwent routine blood tests, visual behavior tests, and magnetic resonance imaging scans before they went to high altitude (Test 1), 7 days after they returned to sea level (Test 2), as well as 3 months (Test 3) after they returned to sea level. In this study, we investigated the hematological parameters, behavioral data, and spontaneous brain activity. There were significant differences among the tests in hematological parameters and spontaneous brain activity. The hematocrit, hemoglobin concentration, and red blood cell count were significantly increased in Test 2 as compared with Tests 1 and 3. As compared with Test 1, Test 3 increased amplitudes of low-frequency fluctuations (ALFF) in the right calcarine gyrus; Tests 2 and 3 increased ALFF in the right supplementary motor cortex, increased regional homogeneity (ReHo) in the left lingual gyrus, increased the voxel-mirrored homotopic connectivity (VMHC) value in the motor cortex, and decreased FC between the left lingual gyrus and left postcentral gyrus. The color accuracy in the visual task was positively correlated with ALFF and ReHo in Test 2. Hypoxia/reoxygenation increased functional connection between the neurons within the visual cortex and the motor cortex but decreased connection between the visual cortex and motor cortex.

Development of new population-averaged standard templates for spatial normalization and segmentation of MR images for postnatal piglet brains.

PROPOSE: To design a set of brain templates for postnatal piglet brains based on high-resolution T1-weighted imaging for voxel-based morphometric analysis. MATERIALS AND METHODS: Using a 3.0 T magnetic resonance (MR) scanner, a population-based whole brain template was developed by averaging forty T1 images in the brains of postnatal piglets at 38 days of age. The templates for gray and white matter, and cerebrospinal fluid were designed based on the corresponding probability maps by adapting individual data sets using statistical parametric mapping. Anatomical labeling maps were generated from labeling propagation derived from the established Pig Brain Atlas. Differences in the coordinates from four significant structural landmarks in the template, plus an additional 12 normalized images and anatomical labeling maps were measured to validate the accuracy of the registration of the template. RESULTS: A whole brain template, a set of tissue-specific probability and anatomical labeling maps were developed. The location deviation of the four significant structural landmarks, including the anterior and posterior regions in the corpus callosum, and the left and right caudate nucleus, was found to be <0.25 cm, validating the sensitivity and resolution of the template. CONCLUSION: A whole brain template map and a set of tissue-specific probability and anatomical labeling maps were developed to analyze the morphometric imaging of the postnatal piglet brain, an animal model of the human infant.