A subject-specific unsupervised deep learning method for quantitative susceptibility mapping using implicit neural representation.

Quantitative susceptibility mapping (QSM) is an MRI-based technique that estimates the underlying tissue magnetic susceptibility based on phase signal. Deep learning (DL)-based methods have shown promise in handling the challenging ill-posed inverse problem for QSM reconstruction. However, they require extensive paired training data that are typically unavailable and suffer from generalization problems. Recent model-incorporated DL approaches also overlook the non-local effect of the tissue phase in applying the source-to-field forward model due to patch-based training constraint, resulting in a discrepancy between the prediction and measurement and subsequently suboptimal QSM reconstruction. This study proposes an unsupervised and subject-specific DL method for QSM reconstruction based on implicit neural representation (INR), referred to as INR-QSM. INR has emerged as a powerful framework for learning a high-quality continuous representation of the signal (image) by exploiting its internal information without training labels. In INR-QSM, the desired susceptibility map is represented as a continuous function of the spatial coordinates, parameterized by a fully-connected neural network. The weights are learned by minimizing a loss function that includes a data fidelity term incorporated by the physical model and regularization terms. Additionally, a novel phase compensation strategy is proposed for the first time to account for the non-local effect of tissue phase in data consistency calculation to make the physical model more accurate. Our experiments show that INR-QSM outperforms traditional established QSM reconstruction methods and the compared unsupervised DL method both qualitatively and quantitatively, and is competitive against supervised DL methods under data perturbations.

Brain Age Prediction Based on Quantitative Susceptibility Mapping Using the Segmentation Transformer.

The process of brain aging is intricate, encompassing significant structural and functional changes, including myelination and iron deposition in the brain. Brain age could act as a quantitative marker to evaluate the degree of the individual's brain evolution. Quantitative susceptibility mapping (QSM) is sensitive to variations in magnetically responsive substances such as iron and myelin, making it a favorable tool for estimating brain age. In this study, we introduce an innovative 3D convolutional network named Segmentation-Transformer-Age-Network (STAN) to predict brain age based on QSM data. STAN employs a two-stage network architecture. The first-stage network learns to extract informative features from the QSM data through segmentation training, while the second-stage network predicts brain age by integrating the global and local features. We collected QSM images from 712 healthy participants, with 548 for training and 164 for testing. The results demonstrate that the proposed method achieved a high accuracy brain age prediction with a mean absolute error (MAE) of 4.124 years and a coefficient of determination (R2) of 0.933. Furthermore, the gaps between the predicted brain age and the chronological age of Parkinson's disease patients were significantly higher than those of healthy subjects (P<0.01). We thus believe that using QSM-based predicted brain age offers a more reliable and accurate phenotype, with the potentiality to serve as a biomarker to explore the process of advanced brain aging.

IMJENSE: Scan-Specific Implicit Representation for Joint Coil Sensitivity and Image Estimation in Parallel MRI.

Parallel imaging is a commonly used technique to accelerate magnetic resonance imaging (MRI) data acquisition. Mathematically, parallel MRI reconstruction can be formulated as an inverse problem relating the sparsely sampled k-space measurements to the desired MRI image. Despite the success of many existing reconstruction algorithms, it remains a challenge to reliably reconstruct a high-quality image from highly reduced k-space measurements. Recently, implicit neural representation has emerged as a powerful paradigm to exploit the internal information and the physics of partially acquired data to generate the desired object. In this study, we introduced IMJENSE, a scan-specific implicit neural representation-based method for improving parallel MRI reconstruction. Specifically, the underlying MRI image and coil sensitivities were modeled as continuous functions of spatial coordinates, parameterized by neural networks and polynomials, respectively. The weights in the networks and coefficients in the polynomials were simultaneously learned directly from sparsely acquired k-space measurements, without fully sampled ground truth data for training. Benefiting from the powerful continuous representation and joint estimation of the MRI image and coil sensitivities, IMJENSE outperforms conventional image or k-space domain reconstruction algorithms. With extremely limited calibration data, IMJENSE is more stable than supervised calibrationless and calibration-based deep-learning methods. Results show that IMJENSE robustly reconstructs the images acquired at 5× and 6× accelerations with only 4 or 8 calibration lines in 2D Cartesian acquisitions, corresponding to 22.0% and 19.5% undersampling rates. The high-quality results and scanning specificity make the proposed method hold the potential for further accelerating the data acquisition of parallel MRI.

APART-QSM: An improved sub-voxel quantitative susceptibility mapping for susceptibility source separation using an iterative data fitting method.

The brain tissue phase contrast in MRI sequences reflects the spatial distributions of multiple substances, such as iron, myelin, calcium, and proteins. These substances with paramagnetic and diamagnetic susceptibilities often colocalize in one voxel in brain regions. Both opposing susceptibilities play vital roles in brain development and neurodegenerative diseases. Conventional QSM methods only provide voxel-averaged susceptibility value and cannot disentangle intravoxel susceptibilities with opposite signs. Advanced susceptibility imaging methods have been recently developed to distinguish the contributions of opposing susceptibility sources for QSM. The basic concept of separating paramagnetic and diamagnetic susceptibility proportions is to include the relaxation rate R2* with R2' in QSM. The magnitude decay kernel, describing the proportionality coefficient between R2' and susceptibility, is an essential reconstruction coefficient for QSM separation methods. In this study, we proposed a more comprehensive complex signal model that describes the relationship between 3D GRE signal and the contributions of paramagnetic and diamagnetic susceptibility to the frequency shift and R2* relaxation. The algorithm is implemented as a constrained minimization problem in which the voxel-wise magnitude decay kernel and sub-voxel susceptibilities are determined alternately in each iteration until convergence. The calculated voxel-wise magnitude decay kernel could realistically model the relationship between the R2' relaxation and the volume susceptibility. Thus, the proposed method effectively prevents the errors of the magnitude decay kernel from propagating to the final susceptibility separation reconstruction. Phantom studies, ex vivo macaque brain experiments, and in vivo human brain imaging studies were conducted to evaluate the ability of the proposed method to distinguish paramagnetic and diamagnetic susceptibility sources. The results demonstrate that the proposed method provides state-of-the-art performances for quantifying brain iron and myelin compared to previous QSM separation methods. Our results show that the proposed method has the potential to simultaneously quantify whole brain iron and myelin during brain development and aging. The proposed model was also deployed with multiple-orientation complex GRE data input measurements, resulting in high-quality QSM separation maps with more faithful tissue delineation between brain structures compared to those reconstructed by single-orientation QSM separation methods.

Iron accumulation and axonal damage of cerebellum in idiopathic cervical dystonia.

BACKGROUND AND PURPOSE: Postmortem brain study indicated that cerebellar Purkinje cell (PC) loss might be a pathological finding in patients with inherited and idiopathic cervical dystonia (ICD). The analysis of conventional magnetic resonance imaging brain scans failed to yield support for this finding. Previous studies have identified that iron overload can be the consequence of neuron death. The objectives of this study were to investigate iron distribution and demonstrate changes in axons in the cerebellum, providing evidence for PC loss in patients with ICD. METHODS: Twenty-eight patients with ICD (20 females) and 28 age- and sex-matched healthy controls were recruited. A spatially unbiased infratentorial template was applied to perform cerebellum optimized quantitative susceptibility mapping and diffusion tensor analysis based on magnetic resonance imaging. Voxel-wise analysis was performed to assess cerebellar tissue magnetic susceptibility and fractional anisotropy (FA) alterations, and the clinical relevance of these findings was investigated in the patients with ICD. RESULTS: Increased susceptibility values revealed by quantitative susceptibility mapping in the right lobule CrusI, CrusII, VIIb, VIIIa, VIIIb and IX were found in the patients with ICD. A reduced FA value was found across almost all the cerebellum; an FA value of the significant clusters within the right lobule VIIIa significantly correlated with the motor severity of patients with ICD (r = -0.575, p = 0.002). CONCLUSIONS: Our study provided evidence for cerebellar iron overload and axonal damage in patients with ICD, which may indicate PC loss and related axonal changes. These results provide evidence for the neuropathological findings in patients with ICD and further highlight the cerebellar involvement in the pathophysiology of dystonia.

An improved asymmetric susceptibility tensor imaging model with frequency offset correction.

PURPOSE: To improve susceptibility tensor imaging (STI) reconstruction using the asymmetric STI model with the correction of non-bulk-magnetic-susceptibility (NBMS) effects. METHOD: A frequency offset term was introduced into the asymmetric STI model to account for the bias between measured MRI frequency signals and conventional susceptibility tensor models because of NBMS contributions. Experiments were conducted to compare the proposed model with conventional STI, conventional STI with the proposed frequency offset correction, and asymmetric STI on simulation, ex vivo mouse brain, and in vivo human brain data. RESULTS: In the simulation where NBMS contributions are head rotation-invariant, the proposed method achieves the lowest errors in mean magnetic susceptibility (MMS) and magnetic susceptibility anisotropy (MSA) and is more robust to noise in the estimation of principal eigenvector (PEV). When considering the head orientation dependency of NBMS contributions, the proposed method shows advantages in estimating MSA and PEV. On the mouse and human brain data, the proposed method produces more reliable MSA maps and more consistent white matter fiber directions when referring to those from DTI than the compared STI methods. CONCLUSION: The proposed method can reduce the effects of NBMS-related frequency shifts on the susceptibility tensors in the brain white matter. This study inspires STI reconstruction from the perspective of better modeling the sources of frequency shifts.

Towards in vivo ground truth susceptibility for single-orientation deep learning QSM: A multi-orientation gradient-echo MRI dataset.

Recently, deep neural networks have shown great potential for solving dipole inversion of quantitative susceptibility mapping (QSM) with improved results. However, these studies utilized their limited dataset for network training and inference, which may lead to untrustworthy conclusions. Thus, a common dataset is needed for a fair comparison between different QSM reconstruction networks. Additionally, finding an in vivo reference susceptibility map that matches acquired single-orientation phase data remains an open problem. Susceptibility tensor imaging (STI) χ33 and Calculation of Susceptibility through Multiple Orientation Sampling (COSMOS) are considered reference susceptibility candidates. However, a large number of multi-orientation GRE data for both STI and COSMOS reconstruction are now unavailable for training supervised neural networks for QSM. In this study, we reported the largest multi-orientation dataset, to the best of our knowledge in the QSM research field, with a total of 144 scans from 8 healthy subjects collected using a 3D GRE sequence from the same MR scanner. In addition, the parcellation of deep gray matter is also provided for automatically extracting susceptibility values. Five recently developed deep neural networks, i.e., xQSM, QSMnet, autoQSM, LPCNN, and MoDL-QSM were performed on this dataset. This potential data source could provide a common framework and labels to test the accuracy and robustness of deep neural networks for QSM reconstruction. This dataset has the potential to provide a benchmark of reference susceptibility for the deep learning-based QSM methods. Additionally, the trained COSMOS-labeled and χ33-labeled networks were tested on the pathological data to explore their potential applications. The data together with deep gray matter parcellation maps are now publicly available via an open repository at https://osf.io/yfms7/, and the raw multi-orientation GRE data are also available at https://osf.io/y6rc3/.

Regularized Asymmetric Susceptibility Tensor Imaging in the Human Brain in Vivo.

Susceptibility tensor imaging (STI) is a promising tool for studying orientation-dependent tissue magnetic susceptibility and for mapping white matter fiber orientations complementary to diffusion tensor imaging (DTI). However, the limited head rotation range within modern head coils for data acquisition makes in vivo STI reconstruction ill-conditioned. Conventional STI reconstruction method is usually vulnerable to noise and requires sufficiently large head rotations to solve this ill-conditioned inverse problem. In this study, based on the recently proposed asymmetric STI (aSTI) model, a new method termed aSTI+ was proposed to improve in vivo STI reconstruction by enforcing isotropic susceptibility tensor inside cerebrospinal fluid (CSF) and applying morphology constraint in white matter. Experimental results showed superior performance of the proposed method with reduced noise, improved tissue contrast and better fiber orientation estimation over previous methods. Thus aSTI+ may promote in vivo human brain STI studies on white matter and myelin-related brain diseases.

MoDL-QSM: Model-based deep learning for quantitative susceptibility mapping.

Quantitative susceptibility mapping (QSM) has demonstrated great potential in quantifying tissue susceptibility in various brain diseases. However, the intrinsic ill-posed inverse problem relating the tissue phase to the underlying susceptibility distribution affects the accuracy for quantifying tissue susceptibility. Recently, deep learning has shown promising results to improve accuracy by reducing the streaking artifacts. However, there exists a mismatch between the observed phase and the theoretical forward phase estimated by the susceptibility label. In this study, we proposed a model-based deep learning architecture that followed the STI (susceptibility tensor imaging) physical model, referred to as MoDL-QSM. Specifically, MoDL-QSM accounts for the relationship between STI-derived phase contrast induced by the susceptibility tensor terms (χ13, χ23 and χ33) and the acquired single-orientation phase. The convolutional neural networks are embedded into the physical model to learn a regularization term containing prior information. χ33 and phase induced by χ13 and χ23 terms were used as the labels for network training. Quantitative evaluation metrics were compared with recently developed deep learning QSM methods. The results showed that MoDL-QSM achieved superior performance, demonstrating its potential for future applications.

Change in Susceptibility Values in Knee Cartilage After Marathon Running Measured Using Quantitative Susceptibility Mapping.

BACKGROUND: Quantitative susceptibility mapping (QSM) has been used to study the magnetic susceptibility properties of collagen fibers in articular cartilage; however, it is unclear whether QSM is sensitive to changes due to degradation caused by long-distance running. It is clinically important to understand the link between long-distance running and microstructural changes in knee cartilage. PURPOSE: To investigate the ability of QSM to assess microstructural changes within cartilage after repetitive loading. STUDY TYPE: Prospective. POPULATION: Thirteen recreational, male long-distance runners. FIELD STRENGTH/SEQUENCE: Three-dimensional gradient recalled echo acquired at 3 T. ASSESSMENT: Magnetic resonance imaging (MRI) and 3D kinematics (translations and rotations during treadmill walking and running) of the knee joint were collected before and after marathon running. The compartments for analysis included the patella, trochlea, and subregions of femoral and tibial cartilage. Changes in regional susceptibility and cartilage thickness were calculated after marathon running. A susceptibility profile was obtained by fitting susceptibility as a function of the normalized depth of cartilage from the superficial to deep layers. STATISTICAL TESTS: Paired t-test or Wilcoxon signed-rank test, 95% confidence interval (CI) of the depth-wise susceptibility profile, Pearson correlation or Spearman correlation. RESULTS: There was a statistically significant increase in susceptibility value in the weight-bearing region of central medial femoral cartilage (cMF-c) after marathon running (pre-marathon: -0.0219 ± 0.0151 ppm, post-marathon: -0.0070 ± 0.0213 ppm, P < 0.05), while the cartilage thickness did not show significant changes in any regions (P-value range: 0.068-0.963). Significant susceptibility elevations occurred in the middle and deep layers of cMF-c (95% CIs did not overlap). A trend toward a positive correlation was found between the changes in susceptibility value in cMF-c and proximal-distal translation of the knee joint during walking (r = 0.55, P = 0.101) and running (r = 0.57, P = 0.089). DATA CONCLUSION: Localized magnetic susceptibility alterations were observed within knee cartilage in the weight-bearing area after repetitive loading without any morphologic changes. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.

Liquid nitrogen waterless fracking for the environmental protection of arid areas during unconventional resource extraction.

Many arid and semi-arid regions are rich in shale gas or coalbed methane. However, hydraulic-fracturing, commonly used for reservoir stimulation, has serious environmental impacts such as the consumption of large quantities of water, damage of residual organic compounds and the disposal of process water. This paper presents liquid nitrogen (LN2) as an environmentally friendly, waterless fracking technology, which could potentially replace hydraulic fracturing. Laboratory experiments on LN2 fracturing were conducted on coal samples, and high-resolution micro X-ray computed tomography was used for 3D visualization and evaluation of fracture evolution characteristics, including liquid nitrogen cyclic quenching, effect of initial fracture size (IFS) and coal saturation. The findings of this study testify to the effectiveness of fracturing by LN2 quenching on coalbed methane reservoirs. This technique would help protect water resources and alleviate other environmental concerns in arid districts during unconventional resource recovery.