Preoperative diffusion tensor imaging: Fiber-trajectory-distribution-based tractography to identify facial nerve in vestibular schwannoma.

PURPOSE: Tractography of the facial nerve based on diffusion MRI is instrumental before surgery for the resection of vestibular schwannoma, but no excellent methods usable for the suppression of motion and image noise have been proposed. The aim of this study was to effectively suppress noise and provide accurate facial nerve reconstruction by extend a fiber trajectory distribution function based on the fourth-order streamline differential equations. METHODS: Preoperative MRI from 33 patients with vestibular schwannoma who underwent surgical resection were utilized in this study. First, T1WI and T2WI were used to obtain mask images and regions of interest. Second, probabilistic tractography was employed to obtain the fibers representing the approximate facial nerve pathway, and these fibers were subsequently translated into orientation information for each voxel. Last, the voxel orientation information and the peaks of the fiber orientation distribution were combined to generate a fiber trajectory distribution function, which was used to parameterize the anatomical information. The parameters were determined by minimizing the cost between the trajectory of fibers and the estimated directions. RESULTS: Qualitative and visual analyses were used to compare facial nerve reconstruction with intraoperative recordings. Compared with other methods (SD_Stream, iFOD1, iFOD2, unscented Kalman filter, parallel transport tractography), the fiber-trajectory-distribution-based tractography provided the most accurate facial nerve reconstructions. CONCLUSION: The fiber-trajectory-distribution-based tractography can effectively suppress the effect of noise. It is a more valuable aid for surgeons before vestibular schwannoma resection, which may ultimately improve the postsurgical patient's outcome.

Investigation into white matter microstructure differences in visual training by using an automated fiber tract subclassification segmentation quantification method.

Visual training has emerged as a useful framework for investigating training-related brain plasticity, a highly complex task involving the interaction of visual orientation, attention, reasoning, and cognitive functions. However, the effects of long-term visual training on microstructural changes within white matter (WM) is poorly understood. Therefore, a set of visual training programs was designed, and automated fiber tract subclassification segmentation quantification based on diffusion magnetic resonance imaging was performed to obtain the anatomical changes in the brains of visual trainees. First, 40 healthy matched participants were randomly assigned to the training group or the control group. The training group underwent 10 consecutive weeks of visual training. Then, the fiber tracts of the subjects were automatically identified and further classified into fiber clusters to determine the differences between the two groups on a detailed scale. Next, each fiber cluster was divided into segments that can analyze specific areas of a fiber cluster. Lastly, the diffusion metrics of the two groups were comparatively analyzed to delineate the effects of visual training on WM microstructure. Our results showed that there were significant differences in the fiber clusters of the cingulate bundle, thalamus frontal, uncinate fasciculus, and corpus callosum between the training group compared and the control group. In addition, the training group exhibited lower mean fractional anisotropy, higher mean diffusivity and radial diffusivity than the control group. Therefore, the long-term cognitive activities, such as visual training, may systematically influence the WM properties of cognition, attention, memory, and processing speed.

Super-resolved q-space learning of diffusion MRI.

BACKGROUND: Diffusion magnetic resonance imaging (dMRI) provides a powerful tool to non-invasively investigate neural structures in the living human brain. Nevertheless, its reconstruction performance on neural structures relies on the number of diffusion gradients in the q-space. High-angular (HA) dMRI requires a long scan time, limiting its use in clinical practice, whereas directly reducing the number of diffusion gradients would lead to the underestimation of neural structures. PURPOSE: We propose a deep compressive sensing-based q-space learning (DCS-qL) approach to estimate HA dMRI from low-angular dMRI. METHODS: In DCS-qL, we design the deep network architecture by unfolding the proximal gradient descent procedure that addresses the compressive sense problem. In addition, we exploit a lifting scheme to design a network structure with reversible transform properties. For implementation, we apply a self-supervised regression to enhance the signal-to-noise ratio of diffusion data. Then, we utilize a semantic information-guided patch-based mapping strategy for feature extraction, which introduces multiple network branches to handle patches with different tissue labels. RESULTS: Experimental results show that the proposed approach can yield a promising performance on the tasks of reconstructed HA dMRI images, microstructural indices of neurite orientation dispersion and density imaging, fiber orientation distribution, and fiber bundle estimation. CONCLUSIONS: The proposed method achieves more accurate neural structures than competing approaches.

Local white matter abnormalities in Parkinson's disease with mild cognitive impairment: Assessed with neurite orientation dispersion and density imaging.

Mild cognitive impairment is a nonmotor complication in Parkinson's disease (PD) that have a high risk of developing dementia. White matter is associated with cognitive function in PD and the alterations may occur before the symptoms of the disease. Previous diffusion tensor imaging (DTI) studies lacked specificity to characterize the concrete contributions of distinct white matter tissue properties. This may lead to inconsistent conclusions about the alteration of white matter microstructure. Here, we used neurite orientation dispersion and density imaging (NODDI) and white matter fiber clustering method to uncover local white matter microstructures in PD with mild cognitive impairment (PD-MCI). This study included 23 PD-MCI and 20 PD with normal cognition (PD-NC) and 21 healthy controls (HC). To probe specific and fine-grained differences, metrics of NODDI and DTI in white matter fiber clusters were evaluated using along-tract analysis. Our results showed that PD-MCI patients had significantly lower neurite density index (NDI) and orientation dispersion index (ODI) in white matter fiber clusters in the prefrontal region. Correlation analysis and receiver operating characteristic (ROC) analysis revealed that the diagnostic performance of NODDI-derived metrics in cingulum bundle (2 clusters) and thalamo-frontal (2 clusters) were superior to DTI metrics. Our study provides a more specific insight to uncover local white matter abnormalities in PD-MCI, which benefit understanding the underlying mechanism of cognitive decline in PD and predicting the disease in advance.

CNTSeg: A multimodal deep-learning-based network for cranial nerves tract segmentation.

The segmentation of cranial nerves (CNs) tracts based on diffusion magnetic resonance imaging (dMRI) provides a valuable quantitative tool for the analysis of the morphology and course of individual CNs. Tractography-based approaches can describe and analyze the anatomical area of CNs by selecting the reference streamlines in combination with ROIs-based (regions-of-interests) or clustering-based. However, due to the slender structure of CNs and the complex anatomical environment, single-modality data based on dMRI cannot provide a complete and accurate description, resulting in low accuracy or even failure of current algorithms in performing individualized CNs segmentation. In this work, we propose a novel multimodal deep-learning-based multi-class network for automated cranial nerves tract segmentation without using tractography, ROI placement or clustering, called CNTSeg. Specifically, we introduced T1w images, fractional anisotropy (FA) images, and fiber orientation distribution function (fODF) peaks into the training data set, and design the back-end fusion module which uses the complementary information of the interphase feature fusion to improve the segmentation performance. CNTSeg has achieved the segmentation of 5 pairs of CNs (i.e. optic nerve CN II, oculomotor nerve CN III, trigeminal nerve CN V, and facial-vestibulocochlear nerve CN VII/VIII). Extensive comparisons and ablation experiments show promising results and are anatomically convincing even for difficult tracts. The code will be openly available at https://github.com/IPIS-XieLei/CNTSeg.

A unified global tractography framework for automatic visual pathway reconstruction.

The human visual pathway starts from the retina, passes through the retinogeniculate visual pathway, the optic radiation, and finally connects to the primary visual cortex. Diffusion MRI tractography is the only technology that can noninvasively reconstruct the visual pathway. However, complete and accurate visual pathway reconstruction is challenging because of the skull base environment and complex fiber geometries. Specifically, the optic nerve within the complex skull base environment can cause abnormal diffusion signals. The crossing and fanning fibers at the optic chiasm, and a sharp turn of Meyer's loop at the optic radiation, contribute to complex fiber geometries of the visual pathway. A fiber trajectory distribution (FTD) function-based tractography method of our previous work and several high sensitivity tractography methods can reveal these complex fiber geometries, but are accompanied by false-positive fibers. Thus, the related studies of the visual pathway mostly applied the expert region of interest selection strategy. However, interobserver variability is an issue in reconstructing an accurate visual pathway. In this paper, we propose a unified global tractography framework to automatically reconstruct the visual pathway. We first extend the FTD function to a high-order streamline differential equation for global trajectory estimation. At the global level, the tractography process is simplified as the estimation of global trajectory distribution coefficients by minimizing the cost between trajectory distribution and the selected directions under the prior guidance by introducing the tractography template as anatomic priors. Furthermore, we use a deep learning-based method and tractography template prior information to automatically generate the mask for tractography. The experimental results demonstrate that our proposed method can successfully reconstruct the visual pathway with high accuracy.

Semi-supervised region-connectivity-based cerebrovascular segmentation for time-of-flight magnetic resonance angiography image.

Deep-learning-based methods have achieved state-of-the-art results in cerebrovascular segmentation. However, it is costly and time-consuming to acquire labeled data because of the complex structure of cerebral vessels. In this paper, we propose a novel semi-supervised cerebrovascular segmentation with a region-connectivity-based mean teacher model (RC-MT) from time-of-flight magnetic resonance angiography (TOF-MRA), where unlabeled data is introduced into the training. Concretely, the RC-MT framework consists of a mean teachers (MT) model and a region-connectivity-based model. The region-connectivity-based model dynamically controls the balance between the supervised loss and unsupervised consistency loss by taking into account that the predicted vessel voxels should be continuous in the underlying anatomy of the brain. Meanwhile, we design a novel multi-scale channel attention fusion Unet (MSCAF-Unet) as a backbone for the student model and the teacher model. The MSCAF-Unet is a multi-scale channel attention fusion layer used to construct an image pyramid input and achieve multi-level receptive field fusion. The proposed method is evaluated on diverse TOF-MRA datasets (three clinical datasets and a public dataset). Experimental results show that the proposed method achieves high-performance gains by incorporating the unlabeled data and outperforms competing semi-supervised-based methods. The code will be openly available at https://github.com/IPIS-XieLei/RC-MT.

Alterations of white matter tracts and topological properties of structural networks in hemifacial spasm.

Hemifacial spasm (HFS) is characterized by involuntary and paroxysmal muscle contractions on the hemiface. It is generally believed that HFS is caused by neurovascular compression at the root exit zone of the facial nerve. In recent years, the structural alterations of brains with HFS have aroused growing concern. However, little attention has been directed towards the possible involvement of specific white matter (WM) tracts and the topological properties of structural networks in HFS. In the present study, diffusion magnetic resonance imaging tractography was utilized to construct structural networks and perform tractometric analysis. The diffusion tensor imaging scalar parameters along with the WM tracts, and the topological parameters of global networks and subnetworks, were assessed in 62 HFS patients and 57 demographically matched healthy controls (HCs). Moreover, we investigated the correlation of these parameters with disease-clinical-level (DCL) and disease-duration-time (DDT) of HFS patients. Compared with HCs, HFS patients had additional hub regions including the amygdala, ventromedial putamen, lateral occipital cortex, and rostral cuneus gyrus. Furthermore, HFS patients showed significant alternations with specific topological properties in some structural subnetworks, including the limbic, default mode, dorsal attention, somato-motor, and control networks, as well as diffusion properties in some WM tracts, including the superior longitudinal fasciculus, cingulum bundle, thalamo-frontal, and corpus callosum. These subnetworks and tracts were associated with the regulation of emotion, motor function, vision, and attention. Notably, we also found that the parameters with subnetworks and tracts exhibited correlations with DCL and DDT. In addition to corroborating previous findings in HFS, this study demonstrates the changed microstructures in specific locations along with the fiber tracts and changed topological properties in structural subnetworks.

Automatic oculomotor nerve identification based on data-driven fiber clustering.

The oculomotor nerve (OCN) is the main motor nerve innervating eye muscles and can be involved in multiple flammatory, compressive, or pathologies. The diffusion magnetic resonance imaging (dMRI) tractography is now widely used to describe the trajectory of the OCN. However, the complex cranial structure leads to difficulties in fiber orientation distribution (FOD) modeling, fiber tracking, and region of interest (ROI) selection. Currently, the identification of OCN relies on expert manual operation, resulting in challenges, such as the carries high clinical, time-consuming, and labor costs. Thus, we propose a method that can automatically identify OCN from dMRI tractography. First, we choose the multi-shell multi-tissue constraint spherical deconvolution (MSMT-CSD) FOD estimation model and deterministic tractography to describe the 3D trajectory of the OCN. Then, we rely on the well-established computational pipeline and anatomical expertise to create a data-driven OCN tractography atlas from 40 HCP data. We identify six clusters belonging to the OCN from the atlas, including the structures of three kinds of positional relationships (pass between, pass through, and go around) with the red nuclei and two kinds of positional relationships with medial longitudinal fasciculus. Finally, we apply the proposed OCN atlas to identify the OCN automatically from 40 new HCP subjects and two patients with brainstem cavernous malformation. In terms of spatial overlap and visualization, experiment results show that the automatically and manually identified OCN fibers are consistent. Our proposed OCN atlas provides an effective tool for identifying OCN by avoiding the traditional selection strategy of ROIs.

Alterations in white matter fiber in Parkinson disease across different cognitive stages.

Cognitive impairment in Parkinson disease (PD) leads to substantial disability. Unlike external manifestations such as tremor, the decay of cognitive function is often an underlying process, and its neuroanatomic substrates are not yet fully elucidated. Knowledge regarding cognitive-related alterations in white matter (WM) pathways helps us understand the mechanisms of cognitive decline in patients with PD. Previous voxel-based analyses with Diffusion tensor imaging (DTI) metrics, such as fractional anisotropy (FA) and mean diffusivity (MD) have uncovered white matter differences in groupwise, but the conclusions were inconsistent. That was partially due to white matter fibers that are known to affect cognition, such as the corpus callosum (CC) and superior longitudinal fasciculus that cross in voxel, and are hard to interpreted by the abovementioned metrics. Furthermore, cognitive decay is a continuous process, it is difficult to reflect the continuous changes of white matter fibers between groups comparison. In the present work, we chose the constrained spherical deconvolution (CSD) and the fixel model, which avoided the effect of crossing fibers. To compare the white matter fiber in different cognitive stages of patients with PD, the results found that the CC, the cingulum bundle (CB), and the corticospinal tract (CST) showed the same trend in the decline of cognitive function, and this change may lead to the impairment of cognitive function. Our findings can help physicians determine the cognitive stage of PD from the perspective of white matter fiber and provide a reference for clinical trials and predictions.

The trajectory of the medial longitudinal fasciculus in the human brain: A diffusion imaging-based tractography study.

The aim of this study is to investigate the trajectory of medial longitudinal fasciculus (MLF) and explore its anatomical relationship with the oculomotor nerve using tractography technique. The MLF and oculomotor nerve were reconstructed at the same time with preset three region of interests (ROIs): one set at the area of rostral midbrain, one placed on the MLF area at the upper pons, and one placed at the cisternal part of the oculomotor nerve. This mapping protocol was tested in an HCP-1065 template, 35 health subjects from Massachusetts General Hospital (MGH), 20 healthy adults and 6 brainstem cavernous malformation (BCM) patients with generalized q-sampling imaging (GQI)-based tractography. Finally, the 200 µm brainstem template from Center for In Vivo Microscopy, Duke University (Duke CIVM), was used to validate the trajectory of reconstructed MLF. The MLF and oculomotor nerve were reconstructed in the HCP-1065 template, 35 MGH health subjects, 20 healthy adults and 6 BCM patients. The MLF was in conjunction with the ipsilateral mesencephalic part of the oculomotor nerve. The displacement of MLF was identified in all BCM patients. Decreased QA, RDI and FA were found in the MLF of lesion side, indicating axonal loss and/or edema of displaced MLF. The reconstructed MLF in Duke CIVM brainstem 200 µm template corresponded well with histological anatomy. The MLF and oculomotor nerve were visualized accurately with our protocol using GQI-based fiber tracking. This GQI-based tractography is an important tool in the reconstruction and evaluation of MLF.

Automated facial-vestibulocochlear nerve complex identification based on data-driven tractography clustering.

Small size and intricate anatomical environment are the main difficulties facing tractography of the facial-vestibulocochlear nerve complex (FVN), and lead to challenges in fiber orientation distribution (FOD) modeling, fiber tracking, region-of-interest selection, and fiber filtering. Experts need rich experience in anatomy and tractography, as well as substantial labor costs, to identify the FVN. Thus, we present a pipeline to identify the FVN automatically, in what we believe is the first study of the automated identification of the FVN. First, we created an FVN template. Forty high-resolution multishell data were used to perform data-driven fiber clustering based on the multishell multitissue constraint spherical deconvolution FOD model and deterministic tractography. We selected the brainstem and cerebellum (BS-CB) region as the seed region and removed the fibers that reach other brain regions. We then performed spectral fiber clustering twice. The first clustering was to create a BS-CB atlas and separate the fibers that pass through the cerebellopontine angle, and the other one was to extract the FVN. Second, we registered the subject-specific fibers in the space of the FVN template and assigned each fiber to the closest cluster to identify the FVN automatically by spectral embedding. We applied the proposed method to different acquirement sites, including two different healthy datasets and two tumor patient datasets. Experimental results showed that our automatic identification results have ideal colocalization with expert manual identification in terms of spatial overlap and visualization. Importantly, we successfully applied our method to tumor patient data. The FVNs identified by the proposed method were in agreement with intraoperative findings.

Investigation of local white matter abnormality in Parkinson's disease by using an automatic fiber tract parcellation.

The deficits of white matter (WM) microstructure are involved during Parkinson's disease (PD) progression. Most current methods identify key WM tracts relying on cortical regions of interest (ROIs). However, such ROI methods can be challenged due to low diffusion anisotropy near the gray matter (GM), which could result in a low sensitivity of tract identification. This work proposes an automatic WM parcellation method to improve the accuracy of WM tract identification and locate abnormal tracts by using sensitive features. The proposed method consists of 1) whole brain WM parcellation using an established fiber clustering method, without using any ROIs, 2) features of fasciculus were calculated to quantify diffusion measures at each equal cross-section along the whole cluster. Then, we use the proposed features to investigate the WM difference in PD compared with healthy controls (HC). We also use these features to investigate the relationship of clinical symptoms and specific fiber tracts. The novelty of the proposed method is that it automatically identifies the abnormal WM fibers in cluster degree. Experiment results indicated that the proposed method had advantage in detecting the local WM abnormality by performing between-group statistical analysis in 30 patients with PD and 28 HC. We found 13 hemisphere clusters and 8 commissural clusters had significant group difference (p < 0.05, corrected by FDR method) in local regions, which belonged to multiple fiber tracts including cingulum bundle (CB), inferior occipito-frontal fasciculus (IoFF), corpus callosum (CC), external capsule (EC), uncinate fasciculus (UF), superior longitudinal fasciculus (SLF) and thalamo front (TF). We also found clusters that had relevance with clinical indices of cognitive function (2 clusters), athletic function (6 clusters), and depressive state (2 clusters) in these significant clusters. From the experiment results, it confirmed the ability of the proposed method to identify potential WM microstructure abnormality.

Local White Matter Fiber Clustering Differentiates Parkinson's Disease Diagnoses.

Scans without evidence of dopaminergic deficit (SWEDD) patients are often misdiagnosed with Parkinson's disease (PD) but have normal dopamine transporter scans. We hypothesised that white matter tracts associated with motor and cognition functions may be affected differently by SWEDD and PD. Automatically annotated fibre clustering (AAFC) is a novel clustering method based on diffusion magnetic resonance imaging (dMRI) tractography that enables highly robust reconstruction of white matter tracts that are composed of corresponding clusters. This study aimed to investigate the white matter properties in the subdivisions of white matter tracts among SWEDD and PD groups. We applied AAFC to identify white matter tracts related to motion and cognition functions in the dataset consisting of SWEDD (n = 22), PD (n = 30) and normal control (NC) (n = 30). Then, we resampled 200 nodes along fibres of cluster, and the diffusion metric values corresponding to each node were calculated and used for comparison. Compared with NC, PD showed significant difference (p < 0.05) in two clusters in thalamo-frontal (TF), one cluster in thalamo-parietal (TP) and one cluster in thalamo-occipital (TO), whereas SWEDD presented no significant difference. Three clusters in cingulum bundle (CB) commonly exhibited significant differences in PD versus SWEDD and NC versus SWEDD. The support vector machine classifier achieved high accuracies in PD-NC, PD-SWEDD and NC-SWEDD classifications. This outcome validated these local white matter differences were useful to separate the three groups. These results suggest that PD exerts more significant effects on thalamo tracts than SWEDD, and unique microstructural changes occur in CB tract in SWEDD.

New Measures for the Coronavirus Disease 2019 Response: A Lesson From the Wenzhou Experience.

As the outbreak of coronavirus disease 2019 (COVID-19) has spread globally, determining how to prevent the spread is of paramount importance. We reported the effectiveness of different responses of 4 affected cities in preventing the COVID-19 spread. We expect the Wenzhou anti-COVID-19 measures may provide information for cities around the world that are experiencing this epidemic.

A ReliefF-SVM-based method for marking dopamine-based disease characteristics: A study on SWEDD and Parkinson's disease.

Parkinson's disease (PD) and scans without evidence of dopaminergic deficit (SWEDD) are two distinct neurological disorders that require different therapeutic approaches; therefore it's critical to classify the two disorders. The neuroimaging technology based on dMRI provided connectivity information and voxel features that can make it possible for researchers to analyze SWEDD and PD differences. In this work, a novel method of ReliefF-SVM-based dMRI analysis was presented to study the potential relations between PD and SWEDD. Some sensorimotor connections were found group-wise differences, and SVM was suggested to successfully classify PD and SWEDD. These results indicate that our method using connectivity information and voxel features may provide a new strategy for disease analysis with small sample data.

Distributed performance of white matter properties in chess players: A DWI study using automated fiber quantification.

The aim of this study is to verify whether anatomical changes occur in the brains of chess players. Besides, it is a potential attempt to evaluate diffusion properties along the tracts due to the diverse situations at anatomical level in different locations; moreover, conventional voxel-based analysis (VBA) has already been used to calculate the average values within the voxels in the investigated regions as analysis data. In this study, we used automated fiber quantification (AFQ) to automatically identify the major tracts that are related to functional domains of the human brain. AFQ can quantify pointwise white matter (WM) properties to detect specific local differences. We selected chess players with superior logical cognition abilities as the carrier to conduct our AFQ experiments. The diffusion properties of the 20 major tracts of professional chess players (n = 28) and matched controls (n = 29) were calculated using diffusion weighted imaging (DWI) data. We noted significant differences (p < 0.05) in the diffusion properties of some successive locations among 100 equidistant points in several tracts, especially in the left superior longitudinal fasciculus(SLF) and right inferior fronto-occipital fasciculus (IFOF). Professional chess players exhibited increase levels in the studied diffusion metrics with Pearson results paralleled the findings. Afterwards, considering the starting and terminating regions of SLF, IFOF, and thalamic radiation, the connectivity of gray matter (GM) where connections are active in the frontal lobe, temporal lobe, and thalamus was assessed to help with the further experiment. The results confirmed the tendency in which anatomical alterations generated different performances along the tracts; furthermore, long-term cognitive activities, such as chess, may systematically influence the WM properties of early memory, attention, and visual pathways.

A Novel Richardson-Lucy Model with Dictionary Basis and Spatial Regularization for Isolating Isotropic Signals.

Diffusion-weighted magnetic resonance imaging is a non-invasive imaging method that has been increasingly used in neuroscience imaging over the last decade. Partial volume effects (PVEs) exist in sampling signal for many physical and actual reasons, which lead to inaccurate fiber imaging. We overcome the influence of PVEs by separating isotropic signal from diffusion-weighted signal, which can provide more accurate estimation of fiber orientations. In this work, we use a novel response function (RF) and the correspondent fiber orientation distribution function (fODF) to construct different signal models, in which case the fODF is represented using dictionary basis function. We then put forward a new index Piso, which is a part of fODF to quantify white and gray matter. The classic Richardson-Lucy (RL) model is usually used in the field of digital image processing to solve the problem of spherical deconvolution caused by highly ill-posed least-squares algorithm. In this case, we propose an innovative model integrating RL model with spatial regularization to settle the suggested double-models, which improve noise resistance and accuracy of imaging. Experimental results of simulated and real data show that the proposal method, which we call iRL, can robustly reconstruct a more accurate fODF and the quantitative index Piso performs better than fractional anisotropy and general fractional anisotropy.