Ze Wang | University of Maryland Baltimore (original) (raw)

Papers by Ze Wang

The human brain exhibits large-scale spontaneous fluctuations that account for most of its total ... more The human brain exhibits large-scale spontaneous fluctuations that account for most of its total energy metabolism. Independent of any overt function, this immense ongoing activity likely creates or maintains a potential functional brain reserve to facilitate normal brain function. An important property of spontaneous brain activity is the long-range temporal coherence, which can be characterized by resting state fMRI-based brain entropy mapping (BEN), a relatively new method that has gained increasing research interest. The purpose of this study was to leverage the large resting state fMRI and behavioral data publicly available from the human connectome project to address three important but still unknown questions: temporal stability of rsfMRI-derived BEN; the relationship of resting BEN to latent functional reserve; associations of resting BEN to neurocognition. Our results showed that rsfMRI-derived BEN was highly stable across time; resting BEN in the default mode network (DMN)...

ArXiv, 2018

Arterial spin labeling perfusion MRI is a noninvasive technique for measuring quantitative cerebr... more Arterial spin labeling perfusion MRI is a noninvasive technique for measuring quantitative cerebral blood flow (CBF), but the measurement is subject to a low signal-to-noise-ratio(SNR). Various post-processing methods have been proposed to denoise ASL MRI but only provide moderate improvement. Deep learning (DL) is an emerging technique that can learn the most representative signal from data without prior modeling which can be highly complex and analytically indescribable. The purpose of this study was to assess whether the record breaking performance of DL can be translated into ASL MRI denoising. We used convolutional neural network (CNN) to build the DL ASL denosing model (DL-ASL) to inherently consider the inter-voxel correlations. To better guide DL-ASL training, we incorporated prior knowledge about ASL MRI: the structural similarity between ASL CBF map and grey matter probability map. A relatively large sample data were used to train the model which was subsequently applied t...

Journal of Alzheimer's disease : JAD, 2012

There is growing recognition of an interaction between cerebrovascular disease and Alzheimer'... more There is growing recognition of an interaction between cerebrovascular disease and Alzheimer's disease, but the mechanisms of this interaction remain poorly understood. While macroscopic stroke can clearly produce cognitive deficits and accelerate Alzheimer's disease, the prevalence and implications of microvascular disease in Alzheimer's disease pathogenesis has been harder to define. At present, white matter (WM) lesions, primarily defined as hyperintensities seen on T2-weighted magnetic resonance imaging (MRI), provide the best biomarker of cerebrovascular disease at the microvascular level. However, T2 hyperintensities in WM can also be caused by other mechanisms such as inflammation. Arterial spin labeled (ASL) perfusion MRI provides a noninvasive approach for quantifying cerebral blood flow (CBF). We explored CBF measurements with ASL in AD patients, mild cognitive impairment patients, and an age-matched control group to determine if CBF in gray matter or WM could ...

NeuroImage, 2013

Arterial Spin Labeling (ASL) can be implemented by combining different labeling schemes and reado... more Arterial Spin Labeling (ASL) can be implemented by combining different labeling schemes and readout sequences. In this study, the performance of 2D and 3D single-shot pulsed-continuous ASL (pCASL) sequences was assessed in a group of young healthy volunteers undergoing a baseline perfusion and a functional study with a sensory-motor activation paradigm. The evaluated sequences were 2D echo-planar imaging (2D EPI), 3D single-shot fast spin echo with in-plane spiral readout (3D FSE spiral), and 3D single-shot gradient-and-spin-echo (3D GRASE). The 3D sequences were implemented with and without the addition of an optimized background suppression (BS) scheme. Labeling efficiency, signal-to-noise ratio (SNR), and gray matter (GM) to white matter (WM) contrast ratio were assessed in baseline perfusion measurements. 3D acquisitions without BS yielded 2-fold increments in spatial SNR, but no change in temporal SNR. The addition of BS to the 3D sequences yielded a 3-fold temporal SNR increase compared to the unsuppressed sequences. 2D EPI provided better GM-to-WM contrast ratio than the 3D sequences. The analysis of functional data at the subject level showed a 3-fold increase in statistical power for the BS 3D sequences, although the improvement was attenuated at the group level. 3D without BS did not increase the maximum t-values, however, it yielded larger activation clusters than 2D. These results demonstrate that BS 3D single-shot imaging sequences improve the performance of pCASL in baseline and activation studies, particularly for individual subject

Magnetic Resonance Imaging, 2006

In independent component analysis (ICA), principal component analysis (PCA) is generally used to ... more In independent component analysis (ICA), principal component analysis (PCA) is generally used to reduce the raw data to a few principal components (PCs) through eigenvector decomposition (EVD) on the data covariance matrix. Although this works for spatial ICA (sICA) on moderately sized fMRI data, it is intractable for temporal ICA (tICA), since typical fMRI data have a high spatial dimension, resulting in an unmanageable data covariance matrix. To solve this problem, two practical data reduction methods are presented in this paper. The first solution is to calculate the PCs of tICA from the PCs of sICA. This approach works well for moderately sized fMRI data; however, it is highly computationally intensive, even intractable, when the number of scans increases. The second solution proposed is to perform PCA decomposition via a cascade recursive least squared (CRLS) network, which provides a uniform data reduction solution for both sICA and tICA. Without the need to calculate the covariance matrix, CRLS extracts PCs directly from the raw data, and the PC extraction can be terminated after computing an arbitrary number of PCs without the need to estimate the whole set of PCs. Moreover, when the whole data set becomes too large to be loaded into the machine memory, CRLS-PCA can save data retrieval time by reading the data once, while the conventional PCA requires numerous data retrieval steps for both covariance matrix calculation and PC extractions. Real fMRI data were used to evaluate the PC extraction precision, computational expense, and memory usage of the presented methods.

Drug and Alcohol Dependence, 2013

Background-Cocaine addicted patients with positive cocaine urine status at treatment entry are fa... more Background-Cocaine addicted patients with positive cocaine urine status at treatment entry are far less likely to have a successful treatment outcome. This work aims to identify brain substrates that can distinguish this group of patients from their cocaine-negative counterparts in order to better understand this clinical phenotype. Going a step beyond conventional functional connectivity, we used wavelet transform coherence (WTC) to determine in which ways the temporal pattern of fMRI cerebral blood flow (CBF) signals during attempted inhibition of cueinduced cocaine craving may differ between these two groups. Methods-Using a critical node in motivational circuitry, amygdala, as a seed, whole brain correlations for the entire sample revealed a functional connection with the dorsal cingulate. Next, WTC maps of CBF were constructed for each individual, characterizing the temporal patterns between these two regions during craving inhibition. Results-As revealed by WTC, during attempted craving inhibition, the cocaine-negative subjects had significantly stronger and longer negative coherence between the amygdala and the dorsal cingulate, as compared to the cocaine-positive subjects. This relationship was neither evident in the resting state nor between two regions unrelated to inhibition processes. Conclusions-The duration and strength of negative coherence calculated from wavelettransformed CBF provide an objective and well-defined way to characterize brain responses during attempted inhibition of cue-induced craving, at the level of the individual. The stronger and sustained negative coherence in CBF between motivational (amygdala) and modulatory (dorsal ☆ Supplementary material can be found by accessing the online version of this paper at http://dx.doi.org. Please see Appendix A for more information.

Frontiers in psychology, 2017

Beneficial effects of repetitive transcranial magnetic stimulation (rTMS) on left dorsolateral pr... more Beneficial effects of repetitive transcranial magnetic stimulation (rTMS) on left dorsolateral prefrontal cortex (DLPFC) have been consistently shown for treating various neuropsychiatrical or neuropsychological disorders, but relatively little is known about its neural mechanisms. Here we conducted a randomized, double-blind, SHAM-controlled study to assess the effects of high-frequency left DLPFC rTMS on resting-state activity. Thirty-eight young healthy subjects received two sessions of either real rTMS (18, 90% motor-threshold; left DLPFC at 20 Hz) or SHAM TMS (20) and functional magnetic resonance imaging scan during rest in 2 days separated by 48 h. Resting-state bran activity was measured with the fractional amplitude of low-frequency fluctuation (fALFF) and functional connectivity (FC). Increased fALFF was found in rostral anterior cingulate cortex (rACC) after 20 Hz rTMS, while no changes were observed after SHAM stimulation. Using the suprathreshold rACC cluster as the see...

Frontiers in psychiatry, 2018

Asthma is a chronic disease appeared to be associated with depression. But the underpinnings of d... more Asthma is a chronic disease appeared to be associated with depression. But the underpinnings of depression in asthma remain unknown. In order to understand the neural mechanisms of depression in asthma, we used cerebral blood flow (CBF) to probe the difference between depressed asthmatic (DA) and non-depressed asthmatic (NDA) patients. Eighteen DA patients, 24 NDA patients and 57 healthy controls (HC) received pulsed arterial spin labeling (pASL) scan for measuring CBF, resting-state functional magnetic resonance imaging (rs-fMRI) scan, severity of depression and asthma control assessment, respectively. Compared to NDA, DA patients showed increased regional CBF (rCBF) in the right cerebellum posterior lobe. Compared to HC, DA, and NDA patients all showed significantly decreased rCBF in the right cerebellum posterior lobe. We showed the first evidence of altered rCBF in the right cerebellum posterior lobe in asthma using pASL, which appeared to be involved in the neuropathology in as...

Frontiers in Aging Neuroscience

Magnetic Resonance in Medicine

Frontiers in human neuroscience, 2018

Chronic smoking impairs brain functions in the prefrontal cortex and the projecting meso-cortical... more Chronic smoking impairs brain functions in the prefrontal cortex and the projecting meso-cortical limbic system. The purpose of this pilot study is to examine whether modulating the frontal brain activity using high-frequency repetitive transcranial magnetic stimulation (rTMS) can improve smoking cessation and to explore the changing pattern of the brain activity after treatment. Fourteen treatment-seeking smokers were offered a program involving 10 days of rTMS treatment with a follow-up for another 25 days. A frequency of 20 Hz rTMS was sequentially applied on the left dorso-lateral prefrontal cortex (DLPFC) and the superior medial frontal cortex (SMFC). The carbon monoxide (CO) level, withdrawal, craving scales, and neuroimaging data were collected. Ten smokers completed the entire treatment program, and 90% of them did not smoke during the 25-day follow-up time. A significant smoking craving reduction and resting brain activity reduction measured by the cerebral blood flow (CBF)...

Medical & biological engineering & computing, 2018

Arterial spin-labeling (ASL) perfusion MRI is a non-invasive method for quantifying cerebral bloo... more Arterial spin-labeling (ASL) perfusion MRI is a non-invasive method for quantifying cerebral blood flow (CBF). Standard ASL CBF calibration mainly relies on pair-wise subtraction of the spin-labeled images and controls images at each voxel separately, ignoring the abundant spatial correlations in ASL data. To address this issue, we previously proposed a multivariate support vector machine (SVM) learning-based algorithm for ASL CBF quantification (SVMASLQ). But the original SVMASLQ was designed to do CBF quantification for all image voxels simultaneously, which is not ideal for considering local signal and noise variations. To fix this problem, we here in this paper extended SVMASLQ into a patch-wise method by using a patch-wise classification kernel. At each voxel, an image patch centered at that voxel was extracted from both the control images and labeled images, which was then input into SVMASLQ to find the corresponding patch of the surrogate perfusion map using a non-linear SVM ...

Scientific reports, Jan 9, 2018

Entropy is an important trait of brain function and high entropy indicates high information proce... more Entropy is an important trait of brain function and high entropy indicates high information processing capacity. We recently demonstrated that brain entropy (BEN) is stable across time and differs between controls and patients with various brain disorders. The purpose of this study was to examine whether BEN is sensitive to pharmaceutical modulations with caffeine. Both cerebral blood flow (CBF) and resting fMRI were collected from sixty caffeine-naïve healthy subjects before and after taking a 200 mg caffeine pill. Our data showed that caffeine reduced CBF in the whole brain but increased BEN across the cerebral cortex with the highest increase in lateral prefrontal cortex, the default mode network (DMN), visual cortex, and motor network, consistent with the beneficial effects of caffeine (such as vigilance and attention) on these areas. BEN increase was correlated to CBF reduction only in several regions (-0.5 < r < -0.4), indicating a neuronal nature for most of the observe...

Oncotarget

and late-onset depression (LOD), and their impact on cognitive function. Thirty-two remitted EOD ... more and late-onset depression (LOD), and their impact on cognitive function. Thirty-two remitted EOD patients, 32 remitted LOD patients, and 43 age-matched healthy controls were recruited, and the pulsed arterial spin labeling data were scanned under 3.0T MRI and processed through voxel-by-voxel statistical analysis. Compared to healthy controls, LOD patients had decreased normalized CBF in the bilateral precuneus, cuneus, right fronto-cingulate-striatal areas, and right temporal, occipital and parietal lobes, but increased normalized CBF in the left frontal and temporal cortices and the cingulate gyrus. EOD patients had decreased normalized CBF in the left cerebellum and right calcarine/lingual/fusiform gyrus, and increased normalized CBF in right angular gyrus. LOD patients displayed hemispheric asymmetry in CBF, and had more regions with abnormal CBF than EOD patients. A significant correlation between abnormal CBF and impaired cognitive function was detected in LOD patients, but not EOD patients. These results demonstrate greater CBF abnormalities in LOD patients than EOD patients, and suggest these CBF changes may be associated with progressive degradation of cognitive function in LOD patients.

PLOS ONE

Arterial Spin Labeled (ASL) perfusion MRI enables non-invasive, quantitative measurements of tiss... more Arterial Spin Labeled (ASL) perfusion MRI enables non-invasive, quantitative measurements of tissue perfusion, and has a broad range of applications including brain functional imaging. However, ASL suffers from low signal-to-noise ratio (SNR), limiting image resolution. Acquisitions using 3D readouts are optimal for background-suppression of static signals, but can be SAR intensive and typically suffer from through-plane blurring. In this study, we investigated the use of accelerated 3D readouts to obtain whole-brain, high-SNR ASL perfusion maps and reduce SAR deposition. Parallel imaging was implemented along the partition-encoding direction in a pseudo-continuous ASL sequence with background-suppression and 3D RARE Stack-Of-Spirals readout, and its performance was evaluated in three small cohorts. First, both non-accelerated and two-fold accelerated single-shot versions of the sequence were evaluated in healthy volunteers during a motor-photic task, and the performance was compared in terms of temporal SNR, GM-WM contrast, and statistical significance of the detected activation. Secondly, single-shot 1D-accelerated imaging was compared to a twoshot accelerated version to assess benefits of SNR and spatial resolution for applications in which temporal resolution is not paramount. Third, the efficacy of this approach in clinical populations was assessed by applying the single-shot 1D-accelerated version to a larger cohort of elderly volunteers. Accelerated data demonstrated the ability to detect functional activation at the subject level, including cerebellar activity, without loss in the perfusion signal temporal stability and the statistical power of the activations. The use of acceleration also resulted in increased GM-WM contrast, likely due to reduced through-plane partial volume effects, that were further attenuated with the use of two-shot readouts. In a clinical cohort, image quality remained excellent, and expected effects of age and sex on cerebral blood flow could be detected. The sequence is freely available upon request for academic use and could benefit a broad range of cognitive and clinical neuroscience research.

Progress in Neuro-Psychopharmacology and Biological Psychiatry

Psychomotor retardation (PMR) is one of the core symptoms of major depressive disorder (MDD) and ... more Psychomotor retardation (PMR) is one of the core symptoms of major depressive disorder (MDD) and has a specific pathophysiology, but studies of PMR remains sparse. The purpose of this study was to explore the cerebral blood flow (CBF) of PMR in MDD. One-hundred-seven antidepressant-free MDD patients and 48 normal controls (NCs) were recruited for this study. All subjects underwent arterial spin labeling-magnetic resonance imaging (ASL-MRI) for the CBF calculation. MDD patients were divided into the PMR group (N=35) and NPMR (non-PMR) group (N=72) according to the Salpetriere Retardation Rating Scale (SRRS) score. After a baseline MRI scan, patients began to receive antidepressant treatment. Thirty-nine patients (15 PMR, 24 NPMR) who were remitted after 8weeks participated in the follow-up MRI scan. For statistical analysis, subjects with unqualified MRI image and unmatched demographic data were ruled out. Consequently, 30 NCs and 60 patients (30 PMR, 30 NPMR) at baseline as well as 22 patients (11 PMR, 11 NPMR) at follow-up underwent statistical analysis. The PMR group showed significantly decreased CBF in the right primary motor cortex (PMC) at baseline, and the CBF value of the right PMC was significantly correlated with the SRRS score, whereas the CBF of the right PMC was increased in the PMR group at follow-up compared with the baseline in longitudinal comparison. Our findings suggest that the CBF of the right PMC is a potential biomarker of PMR in MDD.

European radiology, Jan 24, 2017

We aimed to examine the impact of corticospinal tract (CST) involvement in acute ischaemic stroke... more We aimed to examine the impact of corticospinal tract (CST) involvement in acute ischaemic stroke (AIS) patients on functional outcome and the interaction with reperfusion. We retrospectively examined data in consecutive anterior circulation AIS patients undergoing thrombolysis. MR perfusion (time to maximum of tissue residue function, Tmax) and apparent diffusion coefficient (ADC) images were transformed into standard space and the volumes of CST involvement by Tmax > 6 s (CST-Tmax) and ADC < 620 × 10(-6) mm(2)/s (CST-ADC) lesions were calculated. Good outcome was defined as modified Rankin scale ≤ 2 at 3 months. Reperfusion was defined as a reduction in Tmax > 6 s lesion volume of ≥70% between baseline and 24 h. 82 patients were included. Binary logistic regression revealed that both CST-Tmax and CST-ADC volume at baseline were significantly associated with poor outcome (p < 0.05). The 24-h CST-ADC volume was correlated with baseline CST-ADC volume in patients with rep...

Magnetic Resonance in Medicine, 2016

The goal of this study was to develop a 3D acceleration and reconstruction method to improve imag... more The goal of this study was to develop a 3D acceleration and reconstruction method to improve image quality and resolution of background-suppressed arterial spin-labeled perfusion MRI. Accelerated acquisition was implemented in all three k-space dimensions in a stack-of-spirals readout using variable density spirals and partition undersampling. A single 3D self-consistent parallel imaging (SPIRiT) kernel was calibrated and iteratively applied to reconstruct each imaging volume. Whole-brain (including cerebellum) perfusion imaging was obtained at 3-mm isotropic resolution (nominal) using single- and 2-shot acquisitions and at 2-mm isotropic resolution (nominal) using four-shot acquisitions, achieving effective acceleration factors between 5.5 and 6.6. The signal-to-noise (SNR) performance of 3D SPIRiT was evaluated. The temporal SNR (tSNR) of the cerebral blood flow (CBF) maps and the gray/white matter CBF ratios were quantified. The readout of the arterial spin labeling (ASL) sequence was significantly shortened with acceleration. The CBF values were consistent between accelerated and fully sampled ASL. With shorter spiral interleaves and shorter echo trains, the accelerated images demonstrated reduced blurring and signal dropout in regions with high susceptibility gradients, resulting in improved image quality and increased gray/white matter CBF ratios. The shortened readout was accompanied by a corresponding decrease in tSNR. The 3D acceleration and reconstruction allow a rapid whole-brain readout that improved the quality of ASL perfusion imaging. Magn Reson Med, 2016. © 2016 International Society for Magnetic Resonance in Medicine.

Scientific reports, Jul 5, 2016

Cigarette smoking is a chronic relapsing brain disorder, and remains a premier cause of morbidity... more Cigarette smoking is a chronic relapsing brain disorder, and remains a premier cause of morbidity and mortality. Functional neuroimaging has been used to assess differences in the mean strength of brain activity in smokers' brains, however less is known about the temporal dynamics within smokers' brains. Temporal dynamics is a key feature of a dynamic system such as the brain, and may carry information critical to understanding the brain mechanisms underlying cigarette smoking. We measured the temporal dynamics of brain activity using brain entropy (BEN) mapping and compared BEN between chronic non-deprived smokers and non-smoking controls. Because of the known sex differences in neural and behavioral smoking characteristics, comparisons were also made between males and females. Associations between BEN and smoking related clinical measures were assessed in smokers. Our data showed globally higher BEN in chronic smokers compared to controls. The escalated BEN was associated ...

Medical & biological engineering & computing, Jan 11, 2016

The time-to-time fluctuations (TTFs) of resting-state brain activity as captured by resting-state... more The time-to-time fluctuations (TTFs) of resting-state brain activity as captured by resting-state fMRI (rsfMRI) have been repeatedly shown to be informative of functional brain structures and disease-related alterations. TTFs can be characterized by the mean and the range of successive difference. The former can be measured with the mean squared successive difference (MSSD), which is mathematically similar to standard deviation; the latter can be calculated by the variability of the successive difference (VSD). The purpose of this study was to evaluate both the resting state-MSSD and VSD of rsfMRI regarding their test-retest stability, sensitivity to brain state change, as well as their biological meanings. We hypothesized that MSSD and VSD are reliable in resting brain; both measures are sensitive to brain state changes such as eyes-open compared to eyes-closed condition; both are predictive of age. These hypotheses were tested with three rsfMRI datasets and proven true, suggesting...