Spherical echo-planar time-resolved imaging (sEPTI) for rapid 3D quantitative T 2 * and susceptibility imaging.

PURPOSE: To develop a 3D spherical EPTI (sEPTI) acquisition and a comprehensive reconstruction pipeline for rapid high-quality whole-brain submillimeter T 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification. METHODS: For the sEPTI acquisition, spherical k-space coverage is utilized with variable echo-spacing and maximum kx ramp-sampling to improve efficiency and signal incoherency compared to existing EPTI approaches. For reconstruction, an iterative rank-shrinking B0 estimation and odd-even high-order phase correction algorithms were incorporated into the reconstruction to better mitigate artifacts from field imperfections. A physics-informed unrolled network was utilized to boost the SNR, where 1-mm and 0.75-mm isotropic whole-brain imaging were performed in 45 and 90 s at 3 T, respectively. These protocols were validated through simulations, phantom, and in vivo experiments. Ten healthy subjects were recruited to provide sufficient data for the unrolled network. The entire pipeline was validated on additional five healthy subjects where different EPTI sampling approaches were compared. Two additional pediatric patients with epilepsy were recruited to demonstrate the generalizability of the unrolled reconstruction. RESULTS: sEPTI achieved 1.4 × $$ \times $$ faster imaging with improved image quality and quantitative map precision compared to existing EPTI approaches. The B0 update and the phase correction provide improved reconstruction performance with lower artifacts. The unrolled network boosted the SNR, achieving high-quality T 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification with single average data. High-quality reconstruction was also obtained in the pediatric patients using this network. CONCLUSION: sEPTI achieved whole-brain distortion-free multi-echo imaging and T 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification at 0.75 mm in 90 s which has the potential to be useful for wide clinical applications.

Feasibility of submillimeter functional quantitative susceptibility mapping using 3D echo planar imaging at 7 T.

Quantitative susceptibility mapping (QSM) is a tool for mapping tissue susceptibility. Using QSM for functional brain mapping, it is possible to directly quantify blood-oxygen-level-dependent (BOLD) susceptibility changes. This study presents a submillimeter functional QSM (fQSM) approach compared to BOLD fMRI from data acquired with 3D gradient-echo echo planar imaging (EPI) at ultra-high field. Complex EPI data were acquired in nine healthy subjects with varying temporal and spatial resolutions and used for BOLD fMRI and for fQSM. Right-hand finger tapping experiments were performed as well as one measurement with intentional subject movement. Susceptibility maps were computed using 3D path-based unwrapping, the variable-kernel sophisticated harmonic artifact reduction for phase data, and the streaking artifact reduction for QSM algorithm. Functional data analysis included general linear modeling and computation of z-scores. Submillimeter data were denoised using NOise reduction with DIstribution Corrected (NORDIC), which improved z-scores in the motor cortex for fQSM and fMRI. An expected increase in BOLD fMRI signal and corresponding decrease in magnetic susceptibility was observed in sensorimotor areas during active periods. For all experiments, fQSM showed smaller activation regions compared with fMRI. The percentage of high negative t-values localized in the cortex was higher for fQSM (52%) than for positive or negative t-values for fMRI (45%). For the scans with intentional motion, movement exceeded the size of a voxel, but paradigm dependent signal evolution could be recovered using motion correction. In conclusion, this study demonstrates the feasibility of submillimeter whole-brain fQSM with voxel volume of 0.53 µL. In comparison to traditional BOLD fMRI, fQSM provided improved localization of brain activation within the cortex, especially in submillimeter 3D EPI sequences.

Lifestyle, biological, and genetic factors related to brain iron accumulation across adulthood.

Iron is necessary for many neurobiological mechanisms, but its overaccumulation can be harmful. Factors triggering age-related brain iron accumulation remain largely unknown and longitudinal data are insufficient. We examined associations between brain iron load and accumulation and, blood markers of iron metabolism, cardiovascular health, lifestyle factors (smoking, alcohol use, physical activity, diet), and ApoE status using longitudinal data from the IronAge study (n = 208, age = 20-79, mean follow-up time = 2.75 years). Iron in cortex and basal ganglia was estimated with magnetic resonance imaging using quantitative susceptibility mapping (QSM). Our results showed that (1) higher peripheral iron levels (i.e., composite score of blood iron markers) were related to greater iron load in the basal ganglia; (2) healthier diet was related to higher iron levels in the cortex and basal ganglia, although for the latter the association was significant only in younger adults (age = 20-39); (3) worsening cardiovascular health was related to increased iron accumulation; (4) younger ApoE ε4 carriers accumulated more iron in basal ganglia than younger non-carriers. Our results demonstrate that modifiable factors, including lifestyle, cardiovascular, and physiological ones, are linked to age-related brain iron content and accumulation, contributing novel information on potential targets for interventions in preventing brain iron-overload.

Quantitative susceptibility mapping of subcortical iron deposition in Parkinson disease and multiple system atrophy: clinical correlations and diagnostic implications.

Background: Parkinson disease (PD) and multiple system atrophy (MSA) are neurodegenerative disorders characterized by the accumulation of alpha-synuclein. Distinguishing between these conditions remains a significant challenge. This study thus employed quantitative susceptibility mapping (QSM) to evaluate subcortical iron deposition and its clinical implications in patients with PD or MSA and a group of healthy controls (HCs). Methods: The study included 26 patients with MSA, 40 patients with PD, and 35 HCs. We used magnetic resonance imaging (MRI)-based QSM to measure iron accumulation in the substantia nigra pars compacta (SNc), substantia nigra pars reticulata (SNr), and globus pallidus internus (GPi). We assessed differences between groups, examined correlations with clinical scores, and conducted receiver operating characteristic (ROC) curve analysis. Results: Compared to those with PD, patients with MSA showed more severe motor and nonmotor impairment. QSM analysis indicated a significant increase in iron levels in the SNc, SNr, and GPi regions in patient groups compared to HCs. In patients with MSA, a notable positive correlation was found between SNc QSM values and Non-Motor Symptoms Scale scores (r=0.4; P=0.043). In patients with PD, a positive association was observed between iron levels in the SNc and Unified Parkinson's Disease Rating Scale Part III (UPDRS-III) (r=0.395; P=0.012) and Hamilton Depression Rating Scale scores (r=0.313; P=0.049). Furthermore, iron content in the GPi inversely correlated with rapid-eye movement sleep behavior disorder questionnaire-Hong Kong scores (r=-0.342; P=0.031). The SNr region demonstrated the best ability to discriminate between MSA and PD with an area under the curve (AUC) of 0.67, followed by the GPi (AUC =0.64) and SNc (AUC =0.57). Conclusions: QSM effectively quantified subcortical iron deposition in the PD, MSA, and HC groups. The correlations found between iron levels and clinical manifestations provide insights into the pathophysiological processes of these disorders, highlighting the potential of QSM as a diagnostic tool for differentiation.

Quantitative susceptibility mapping based basal ganglia segmentation via AGSeg: leveraging active gradient guiding mechanism in deep learning.

Background: With better visual contrast and the ability for magnetic susceptibility quantification analysis, quantitative susceptibility mapping (QSM) has emerged as an important magnetic resonance imaging (MRI) method for basal ganglia studies. Precise segmentation of basal ganglia is a prerequisite for quantification analysis of tissue magnetic susceptibility, which is crucial for subsequent disease diagnosis and surgical planning. The conventional method of localizing and segmenting basal ganglia heavily relies on layer-by-layer manual annotation by experts, resulting in a tedious amount of workload. Although several morphology registration and deep learning based methods have been developed to automate segmentation, the voxels around the nuclei boundary remain a challenge to distinguish due to insufficient tissue contrast. This paper proposes AGSeg, an active gradient guidance-based susceptibility and magnitude information complete (MIC) network for real-time and accurate basal ganglia segmentation. Methods: Various datasets, including clinical scans and data from healthy volunteers, were collected across multiple centers with different magnetic field strengths (3T/5T/7T), with a total of 210 three-dimensional (3D) susceptibility measurements. Manual segmentations following fixed rules for anatomical borders annotated by experts were used as ground truth labels. The proposed network took QSM maps and Magnitude images as two individual inputs, of which the features are selectively enhanced in the proposed magnitude information complete (MIC) module. AGSeg utilized a dual-branch architecture, with Seg-branch aiming to generate a proper segmentation map and Grad-branch to reconstruct the gradient map of regions of interest (ROIs). With the support of the newly designed active gradient module (AGM) and gradient guiding module (GGM), the Grad-branch provided attention guidance for the Seg-branch, facilitating it to focus on the boundary of target nuclei. Results: Ablation studies were conducted to assess the functionality of the proposed modules. Significant performance decrement was observed after ablating relative modules. AGSeg was evaluated against several existing methods on both healthy and clinical data, achieving an average Dice similarity coefficient (DSC) =0.874 and average 95% Hausdorff distance (HD95) =2.009. Comparison experiments indicated that our model had superior performance on basal ganglia segmentation and better generalization ability over existing methods. The AGSeg outperformed all implemented comparison deep learning algorithms with average DSC enhancement ranging from 0.036 to 0.074. Conclusions: The current work integrates a deep learning-based method into automated basal ganglia segmentation. The high processing speed and segmentation robustness of AGSeg contribute to the feasibility of future surgery planning and intraoperative navigation. Experiments show that leveraging active gradient guidance mechanisms and magnitude information completion can facilitate the segmentation process. Moreover, this approach also offers a portable solution for other multi-modality medical image segmentation tasks.

Comparative analysis of quantitative susceptibility mapping in preclinical dementia detection.

PURPOSE: This review aims to explore the role of Quantitative Susceptibility Mapping (QSM) in the early detection of neurodegenerative diseases, particularly Alzheimer's disease (AD) and Lewy body dementia (LBD). By examining QSM's ability to map brain iron deposition, we seek to highlight its potential as a diagnostic tool for preclinical dementia. METHODOLOGY: QSM techniques involve the advanced processing of MRI phase images to reconstruct tissue susceptibility, employing methods such as spherical mean value filtering and Tikhonov regularization for accurate background field removal. This review discusses how these methodologies enable the precise quantification of iron and other elements within the brain. RESULTS: QSM has demonstrated effectiveness in identifying early pathological changes in key brain regions, including the hippocampus, basal ganglia, and substantia nigra. These regions are significantly impacted in the early stages of AD and LBD. Studies reviewed indicate that QSM can detect subtle neurodegenerative changes, providing valuable insights into disease progression. However, challenges remain in standardizing QSM processing algorithms to ensure consistent results across different studies. CONCLUSION: QSM emerges as a promising tool for early dementia detection, offering precise measurements of brain iron deposition and other critical biomarkers. The review underscores the importance of refining QSM methodologies and integrating them with other imaging modalities to improve early diagnosis and management of neurodegenerative diseases. Future research should focus on standardizing QSM techniques and exploring their synergistic use with other neuroimaging methods to enhance its clinical utility.

Plug-and-Play latent feature editing for orientation-adaptive quantitative susceptibility mapping neural networks.

Quantitative susceptibility mapping (QSM) is a post-processing technique for deriving tissue magnetic susceptibility distribution from MRI phase measurements. Deep learning (DL) algorithms hold great potential for solving the ill-posed QSM reconstruction problem. However, a significant challenge facing current DL-QSM approaches is their limited adaptability to magnetic dipole field orientation variations during training and testing. In this work, we propose a novel Orientation-Adaptive Latent Feature Editing (OA-LFE) module to learn the encoding of acquisition orientation vectors and seamlessly integrate them into the latent features of deep networks. Importantly, it can be directly Plug-and-Play (PnP) into various existing DL-QSM architectures, enabling reconstructions of QSM from arbitrary magnetic dipole orientations. Its effectiveness is demonstrated by combining the OA-LFE module into our previously proposed phase-to-susceptibility single-step instant QSM (iQSM) network, which was initially tailored for pure-axial acquisitions. The proposed OA-LFE-empowered iQSM, which we refer to as iQSM+, is trained in a simulated-supervised manner on a specially-designed simulation brain dataset. Comprehensive experiments are conducted on simulated and in vivo human brain datasets, encompassing subjects ranging from healthy individuals to those with pathological conditions. These experiments involve various MRI platforms (3T and 7T) and aim to compare our proposed iQSM+ against several established QSM reconstruction frameworks, including the original iQSM. The iQSM+ yields QSM images with significantly improved accuracies and mitigates artifacts, surpassing other state-of-the-art DL-QSM algorithms. The PnP OA-LFE module's versatility was further demonstrated by its successful application to xQSM, a distinct DL-QSM network for dipole inversion. In conclusion, this work introduces a new DL paradigm, allowing researchers to develop innovative QSM methods without requiring a complete overhaul of their existing architectures.

Quantitative susceptibility mapping through model-based deep image prior (MoDIP).

The data-driven approach of supervised learning methods has limited applicability in solving dipole inversion in Quantitative Susceptibility Mapping (QSM) with varying scan parameters across different objects. To address this generalization issue in supervised QSM methods, we propose a novel training-free model-based unsupervised method called MoDIP (Model-based Deep Image Prior). MoDIP comprises a small, untrained network and a Data Fidelity Optimization (DFO) module. The network converges to an interim state, acting as an implicit prior for image regularization, while the optimization process enforces the physical model of QSM dipole inversion. Experimental results demonstrate MoDIP's excellent generalizability in solving QSM dipole inversion across different scan parameters. It exhibits robustness against pathological brain QSM, achieving over 32 % accuracy improvement than supervised deep learning methods. It is also 33 % more computationally efficient and runs 4 times faster than conventional DIP-based approaches, enabling 3D high-resolution image reconstruction in under 4.5 min.

Harmonic field extension for QSM with reduced spatial coverage using physics-informed generative adversarial network.

Quantitative susceptibility mapping (QSM) is frequently employed in investigating brain iron related to brain development and diseases within deep gray matter (DGM). Nonetheless, the acquisition of whole-brain QSM data is time-intensive. An alternative approach, focusing the QSM specifically on areas of interest such as the DGM by reducing the field-of-view (FOV), can significantly decrease scan times. However, severe susceptibility value underestimations have been reported during QSM reconstruction with a limited FOV, largely attributable to artifacts from incorrect background field removal in the boundary region. This presents a considerable barrier to the clinical use of QSM with small spatial coverages using conventional methods alone. To mitigate the propagation of these errors, we proposed a harmonic field extension method based on a physics-informed generative adversarial network. Both quantitative and qualitative results demonstrate that our method outperforms conventional methods and delivers results comparable to those obtained with full FOV. Furthermore, we demonstrate the versatility of our method by applying it to data acquired prospectively with limited FOV and to data from patients with Parkinson's disease. The method has shown significant improvements in local field results, with QSM outcomes. In a clear illustration of its feasibility and effectiveness in real clinical environments, our proposed method addresses the prevalent issue of susceptibility underestimation in QSM with small spatial coverage.

Quantitative susceptibility mapping from basal ganglia and related structures: correlation with disease severity in progressive supranuclear palsy.

OBJECTIVE: We examined whether mean magnetic susceptibility values from deep gray matter structures in patients with progressive supranuclear palsy (PSP) differed from those in patients with Parkinson's disease (PD) and healthy volunteers, and correlated with the PSP rating scale. METHODS: Head of caudate nucleus, putamen, globus pallidus, substantia nigra and red nucleus were the regions of interest. Mean susceptibility values from these regions in PSP patients were estimated using quantitative susceptibility mapping. Correlations with clinical severity of disease as measured by the PSP rating scale were examined. The mean susceptibility values were also compared with those from healthy volunteers and age- and disease duration-matched patients with PD. RESULTS: Data from 26 healthy volunteers, 26 patients with PD and 27 patients with PSP, were analysed. Patients with PSP had higher mean susceptibility values from all regions of interest when compared to both the other groups. The PSP rating scale scores correlated strongly with mean susceptibility values from the red nucleus and moderately with those from the putamen and substantia nigra. The scores did not correlate with mean susceptibility values from the caudate nucleus or globus pallidus. In patients with PD, the motor deficits correlated moderately with mean susceptibility values from substantia nigra. CONCLUSIONS: In patients with PSP, mean susceptibility values indicating the severity of mineralization of basal ganglia and related structures correlate with disease severity, the correlation of red nucleus being the strongest. Further studies are warranted to explore whether mean susceptibility values could serve as biomarkers for PSP.

Histogram analysis of quantitative susceptibility mapping and apparent diffusion coefficient for identifying isocitrate dehydrogenase genotypes and tumor subtypes of adult-type diffuse gliomas.

Background: Accurate preoperative identification of isocitrate dehydrogenase (IDH) genotypes and tumor subtypes is highly important for proper treatment planning and prognosis evaluation in patients with glioma. This study aimed to differentiate IDH genotypes and tumor subtypes of adult-type diffuse gliomas using histogram features of quantitative susceptibility mapping (QSM) and apparent diffusion coefficient (ADC). Methods: This prospective study enrolled patients with suspected gliomas between March 2019 and January 2022 in a random series. Histogram features of QSM and ADC were extracted from the tumor parenchyma. The Mann-Whitney U test was used to compare the difference in histogram features between different IDH genotypes and among tumor subtypes. Receiver operating characteristic (ROC) curves were constructed to assess the corresponding diagnostic performance. Results: This study included 47 patients with histopathologically confirmed adult-type diffuse gliomas. Totals of seven QSM features including 10th percentile (P10), 90th percentile (P90), interquartile range (IQR), maximum, mean absolute deviation (MAD), root mean squared (RMS), and variance, and five ADC features including P10, mean, median, RMS, and skewness exhibited significant differences between different IDH genotypes (P<0.05 for all), with the IQR of QSM demonstrating the highest area under curve (AUC) of 0.774 [95% confidence interval (CI): 0.635-0.913]. For separating tumor subtypes, the IQR of QSM also showed the highest AUC of 0.745 (95% CI: 0.566-0.924) for glioblastoma (GBM) versus astrocytoma and 0.848 (95% CI: 0.706-0.989) for GBM versus oligodendroglioma, but none of the features could discriminate astrocytoma from oligodendroglioma. The combination of the IQR of QSM, P10 of ADC, and age achieved the highest AUC of 0.910 (95% CI: 0.826-0.994) for IDH genotypes, and 0.939 (95% CI: 0.859-1.000) and 0.967 (95% CI: 0.904-1.000) for GBM versus astrocytoma and GBM versus oligodendroglioma, respectively. Conclusions: QSM and ADC histogram features may serve as potential imaging markers for noninvasively assessing IDH genotypes and tumor subtypes of adult-type diffuse gliomas. Combining significant features may enhance the diagnostic performance substantially.

Correlation of brain iron deposition and freezing of gait in Parkinson's disease: a cross-sectional study.

Background: Quantitative susceptibility mapping (QSM) is a novel imaging method for detecting iron content in the brain. The study aimed determine whether the iron deposition in the brains of people with Parkinson's disease (PD) is correlated with freezing of gait (FOG). Methods: We retrospectively collected the data of 24 patients with PD from the Movement Disorders Program and 36 healthy controls (HCs) from January 2021 to December 2021. Clinical assessments included mental intelligence scales, Parkinson rating scales, motor-related scales, and clinical gait assessments. All exercise scales and gait assessments were performed in the "ON" and "OFF" states. Magnetic resonance imaging (MRI) data were collected using 3-dimensional fast low-angle shot sequences. We chose the bilateral red nucleus, substantia nigra, thalamus, putamen, caudate nucleus, and globus pallidus as regions of interest for QSM analysis. Results: The iron deposition in the substantia nigra of the PD group was significantly higher than that of the HC group (P<0.01). In the PD group, the iron deposition in the substantia nigra of patients with FOG was significantly higher than that in patients without FOG (P=0.04). The iron deposition in the substantia nigra was positively correlated with the New Freezing of Gait Questionnaire (P=0.03). The scores for depression and anxiety of the PD group were significantly higher than those of the HC group, while the Berg balance scale score was significantly lower (P<0.01). Conclusions: The iron deposition in the substantia nigra of patients with PD is increased compared with that of controls and is associated with FOG. QSM can be used to detect brain iron deposition in patients with PD, which would help to explore the mechanism of abnormal neurobiological activity in FOG.

Evaluating the role of Ginkgo biloba extract in the secondary prevention of acute ischemic stroke with cerebral microbleeds by quantitative susceptibility mapping (QSM).

BACKGROUND: EGb 761, a standardized dry extract of Ginkgo biloba leaves, has certain anti-inflammatory and thrombotic effects and can be used to treat cerebrovascular diseases. METHODS: A total of 49 patients were randomly assigned to the Aspirin group (24 cases in Controlled group) and the Aspirin + Ginkgo biloba group (25 cases in Treatment group). The quantitative magnetic sensitivity and venous oxygen saturation of cerebral microbleeds were analyzed at admission, discharge, and after follow-up for 3 and 6 months. RESULTS: The demographic details age, gender, and admission to NIHSS were not significantly different between the two groups (p < 0.05). Quantitative susceptibility mapping (QSM) showed that the magnetic sensitivity of patients in both groups remained stable after 3 and 6 months of follow-up, while the venous oxygen saturation of the Treatment group increased. The venous oxygen saturation at 3 and 6 months of follow-up was negatively correlated with the modified mRS grade score. CONCLUSIONS: QSM can be used as a quantitative follow-up tool in monitoring both oxygen saturation and Magnetic susceptibility of microbleeds noninvasively in ischemic stroke patients with cerebral microbleeds. EGB combined with Aspirin can improve blood oxygen saturation in those patients and this effect is particularly significant in the long-term efficacy of secondary prevention.

Effect of corticosubcortical iron deposition on dysfunction in CADASIL is mediated by white matter microstructural damage.

Iron dysregulation may attenuate cognitive performance in patients with CADASIL. However, the underlying pathophysiological mechanisms remain incompletely understood. Whether white matter microstructural changes mediate these processes is largely unclear. In the present study, 30 cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) patients were confirmed via genetic analysis and 30 sex- and age-matched healthy controls underwent multimodal MRI examinations and neuropsychological assessments. Quantitative susceptibility mapping and peak width of skeletonized mean diffusivity (PSMD) were analyzed. Mediation effect analysis was performed to explore the interrelationship between iron deposition, white matter microstructural changes and cognitive deficits in CADASIL. Cognitive deterioration was most affected in memory and executive function, followed by attention and working memory in CADASIL. Excessive iron in the temporal-precuneus pathway and deep gray matter specific to CADASIL were identified. Mediation analysis further revealed that PSMD mediated the relationship between iron concentration and cognitive profile in CADASIL. The present findings provide a new perspective on iron deposition in the corticosubcortical circuit and its contribution to disease-related selective cognitive decline, in which iron concentration may affect cognition by white matter microstructural changes in CADASIL.

Thin slab quantitative susceptibility mapping.

PURPOSE: Susceptibility maps reconstructed from thin slabs may suffer underestimation due to background-field removal imperfections near slab boundaries and the increased difficulty of solving a 3D-inversion problem with reduced support, particularly in the direction of the main magnetic field. Reliable QSM reconstruction from thin slabs would enable focal acquisitions in a much-reduced scan time. METHODS: This work proposes using additional rapid low-resolution data of extended spatial coverage to improve background-field estimation and regularize the inversion-to-susceptibility process for high resolution, thin slab data. The new method was tested using simulated and in-vivo brain data of high resolution (0.33 × 0.33 × 0.33 mm3 and 0.54 × 0.54 × 0.65 mm3 , respectively) at 3T, and compared to the standard large volume approach. RESULTS: Using the proposed method, in-vivo high-resolution QSM at 3T was obtained from slabs of width as small as 10.4 mm, aided by a lower-resolution dataset of 24 times coarser voxels. Simulations showed that the proposed method produced more consistent measurements from slabs of at least eight slices. Reducing the mean ROI error to 5% required the low-resolution data to cover ˜60 mm in the direction of the main field, have at least 2-mm isotropic resolution that is not coarser than the high-resolution data by more than four-fold in any direction. CONCLUSION: Applying the proposed method enabled focal QSM acquisitions at sub-millimeter resolution within reasonable acquisition time.

Evaluation of iron deposition in the motor CSTC loop of a Chinese family with paroxysmal kinesigenic dyskinesia using quantitative susceptibility mapping.

Introduction: Previous studies have revealed structural, functional, and metabolic changes in brain regions inside the cortico-striatal-thalamo-cortical (CSTC) loop in patients with paroxysmal kinesigenic dyskinesia (PKD), whereas no quantitative susceptibility mapping (QSM)-related studies have explored brain iron deposition in these areas. Methods: A total of eight familial PKD patients and 10 of their healthy family members (normal controls) were recruited and underwent QSM on a 3T magnetic resonance imaging system. Magnetic susceptibility maps were reconstructed using a multi-scale dipole inversion algorithm. Thereafter, we specifically analyzed changes in local mean susceptibility values in cortical regions and subcortical nuclei inside the motor CSTC loop. Results: Compared with normal controls, PKD patients had altered brain iron levels. In the cortical gray matter area involved with the motor CSTC loop, susceptibility values were generally elevated, especially in the bilateral M1 and PMv regions. In the subcortical nuclei regions involved with the motor CSTC loop, susceptibility values were generally lower, especially in the bilateral substantia nigra regions. Conclusion: Our results provide new evidence for the neuropathogenesis of PKD and suggest that an imbalance in brain iron levels may play a role in PKD.

Automated detection of cerebral microbleeds on MR images using knowledge distillation framework.

Introduction: Cerebral microbleeds (CMBs) are associated with white matter damage, and various neurodegenerative and cerebrovascular diseases. CMBs occur as small, circular hypointense lesions on T2*-weighted gradient recalled echo (GRE) and susceptibility-weighted imaging (SWI) images, and hyperintense on quantitative susceptibility mapping (QSM) images due to their paramagnetic nature. Accurate automated detection of CMBs would help to determine quantitative imaging biomarkers (e.g., CMB count) on large datasets. In this work, we propose a fully automated, deep learning-based, 3-step algorithm, using structural and anatomical properties of CMBs from any single input image modality (e.g., GRE/SWI/QSM) for their accurate detections. Methods: In our method, the first step consists of an initial candidate detection step that detects CMBs with high sensitivity. In the second step, candidate discrimination step is performed using a knowledge distillation framework, with a multi-tasking teacher network that guides the student network to classify CMB and non-CMB instances in an offline manner. Finally, a morphological clean-up step further reduces false positives using anatomical constraints. We used four datasets consisting of different modalities specified above, acquired using various protocols and with a variety of pathological and demographic characteristics. Results: On cross-validation within datasets, our method achieved a cluster-wise true positive rate (TPR) of over 90% with an average of <2 false positives per subject. The knowledge distillation framework improves the cluster-wise TPR of the student model by 15%. Our method is flexible in terms of the input modality and provides comparable cluster-wise TPR and better cluster-wise precision compared to existing state-of-the-art methods. When evaluating across different datasets, our method showed good generalizability with a cluster-wise TPR >80 % with different modalities. The python implementation of the proposed method is openly available.