HLHanzhang Lu

Hanzhang Lu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hanzhang Lu has a total of 600 co-authors. Here are the top co-authors by number of shared publications:

  • Peiying Liu

    Department of Diagnostic Radiology & Nuclear Medicine University of Maryland School of Medicine Baltimore Maryland USA

    39 shared publications
  • Peiying Liu

    Department of Radiology Johns Hopkins University School of Medicine Baltimore Maryland USA

    26 shared publications
  • Yang Li

    Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA

    19 shared publications
  • Zixuan Lin

    Department of Radiology Johns Hopkins University School of Medicine Baltimore Maryland USA

    19 shared publications
  • Dengrong Jiang

    National Engineering Research Center of Advanced Magnetic Resonance Technologies for Diagnosis and Therapy, School of Biomedical Engineering Shanghai Jiao Tong University Shanghai China

    18 shared publications
  • Dengrong Jiang

    Department of Radiology Johns Hopkins University School of Medicine Baltimore Maryland USA

    15 shared publications

Latest Publications

  • Open Access

    Cerebral oxygen extraction fraction as a physiological biomarker in vascular cognitive impairment: A study in older adults enriched for small vessel diseases

    by Jie Song, Wen Shi, Julia Suconic, Kaisha Hazel, George Pottanat, Ebony Jones, Tyler Deane, Archana Sharma, Cuimei Xu, Kumiko Oishi, et al.

    INTRODUCTION: Vascular cognitive impairment and dementia (VCID) is a common cause of cognitive impairment, yet validated biomarkers, particularly those reflecting vascular function, remain lacking. Oxygen extraction fraction (OEF), which reflects the functional status of cerebral small vessels, is a potential physiological biomarker for VCID. We tested this hypothesis in older adults enriched for small vessel diseases, controlling for Alzheimer's disease (AD) co‐pathology. METHODS: Comprehensive cognitive, plasma, and imaging data were collected from 81 older adults with at least one vascular risk factor. OEF in major cerebral veins was quantified using magnetic resonance imaging. Associations between OEF and cognition, vascular, and plasma AD biomarkers were evaluated. RESULTS: In this VCID‐enriched cohort, poorer global cognition and executive function were associated with elevated OEF, but not with plasma phosphorylated tau 217. Increased OEF correlated with more vascular risk factors and greater white matter hyperintensity volume. DISCUSSION: Our findings suggest that elevated OEF in individuals with vascular risks is a physiological biomarker for VCID.


  • Open Access

    Five‐year change in brain metabolism across the spectrum of cognitive impairment in older adults: a quantitative MRI study

    by Jiani Wu, Kumiko Oishi, Anja Soldan, Corinne Pettigrew, Zixuan Lin, Yuxin Zhu, Dengrong Jiang, Xin Li, Yifan Gou, Abhay Moghekar, et al.

    INTRODUCTION: It is desirable to measure brain metabolism without needing ionizing radiation and elucidate its time trajectory. This study employed a positron emission tomography ‐validated, non‐invasive magnetic resonance imaging (MRI) technique to characterize longitudinal changes in cerebral metabolic rate of oxygen (CMRO 2 ) in both cognitively normal and impaired older adults. METHODS: Participants received serial MRI‐based CMRO 2 scans between 2015 and 2025. Levels of AD pathology were measured in cerebrospinal fluid and plasma at the initial MRI. RESULTS: At the initial visit, participants with MCI/dementia had lower CMRO 2 than normal participants. In the longitudinal analyses, CMRO 2 decreased over time in those who were cognitively impaired but remained unchanged in normals. Among normal participants, higher initial levels of total tau and phosphorylated tau at threonine 181 were associated with an increase in CMRO 2 during follow‐up, especially in amyloid‐positive individuals. DISCUSSION: Aside from the well‐known metabolic slowdown in the later phases of AD, the brain exhibits tau‐dependent hypermetabolism during the preclinical stage.

  • Open Access

    Imaging Venous Transit Time and Blood–Brain Barrier Permeability of the Deep Brain

    by Wen Shi, Jie Song, Jiani Wu, Yifan Gou, Zhiyi Hu, Vivek Yedavalli, Zixuan Lin, Hanzhang Lu

    Purpose: Imaging of venous vessels with advanced pulse sequences allows the assessment of important hemodynamic parameters. Previous work has measured venous transit time (VTT) and blood–brain barrier (BBB) water permeability in cortical tissue. This study aimed to measure VTT and BBB permeability in the deep brain through imaging in the straight sinus (SS). Methods: Two studies were conducted. Study I measured VTT of the deep brain using a Venous‐transit‐time‐Imaging‐by‐Changes‐in‐T1‐Relaxation (VICTR) MRI sequence. The obtained VTT was used in study II to optimize the imaging parameters of Water‐Extraction‐with‐Phase‐Contrast‐Arterial‐Spin‐Tagging (WEPCAST) MRI pulse sequence, and the permeability‐surface‐area product (PS) was quantified in the deep brain. Test–retest reproducibility was assessed. Results: SS VTT was found to be 2807.5 ± 349.5 ms, which was approximately half a second shorter than the SSS VTT (3365.0 ± 235.3 ms, p < 0.001). The SS VTT results were reproducible across repetitions, with a coefficient of variation of 2.8% ± 0.5%. The standard deviation of the VTT distribution in SS was 973.7 ± 48.1 ms, again shorter than the corresponding SSS values (1165.5 ± 37.2 ms, p < 0.001). The water extraction fraction, E , from the SS was 94.8% ± 1.9%, and the deep‐brain PS was 141.3 ± 20.2 mL/100 g/min. The PS values were slightly lower than the cortical PS of 152.9 ± 20.9 mL/100 g/min ( p = 0.039), and may reflect the lower vascularity in the deep‐brain region. The intraclass correlation coefficient (ICC) of the deep‐brain PS values from repeated runs was 0.863. Conclusion: This is the first study to measure two parameters associated with deep‐brain physiology, specifically VTT and BBB water permeability, and provide their normative values.


  • Open Access

    MRI‐based biomarkers of cerebrovascular function and their role in VCID

    by Hanzhang Lu

    Background: MRI provides an excellent tool in assessing cerebrovascular function and presents an important opportunity for the development of biomarkers in vascular contributions to cognitive impairment and dementia (VCID). In this work, we will review recent development in MRI‐based biomarkers of cerebrovascular physiology and their potential role in VCID. Method: This work will discuss 3 important parameters related to cerebrovascular function, measured with MRI (Figure 1). Section 1 will describe cerebral blood flow (CBF). CBF (Figure 1a) of the brain can be measured with a non‐contrast MRI technique referred to as arterial spin labeling (ASL). Section 2 will discuss oxygen extraction fraction (OEF). OEF (Figure 1b) reflects the balance between oxygen supply and consumption and is a known hallmark of ischemia. Section 3 will discuss a cerebrovascular reactivity (CVR) measure that reflects the brain's dynamic vascular function of vasodilatory capacity. CVR (Figure 1b) has been evaluated in the MarkVCID Consortium. Result: CBF has been investigated extensively in the context of Alzheimer's dementia. The general findings were that CBF was diminished in default model network brain regions such as posterior cingulate cortex and angular gyrus. This is thought to be related to reduced neural activity, as opposed to impaired vascular function. CBF alterations in VCID is less studied. A few studies including sporadic and genetic small vessel disease have suggested that CBF in VCID is reduced in a global fashion. OEF studies suggest that the brain's OEF is differentially affected in AD and VCID. With AD, OEF was often found to be diminished, presumably due to neurodegeneration and reduced brain metabolic rate. With VCID, OEF is elevated with higher vascular risk factors (Figure 2a) and increased faster longitudinally (Figure 2b). CVR is diminished in impaired individuals and is most strongly correlated with the Montreal Cognitive Assessment (MoCA) score (Figure 3). This association was found to be independent of AD pathological measures such as a‐beta42, tau, and ptau. CVR was also associated with executive function. Conclusion: Several MRI‐based cerebrovascular measures have shown strong promises as biomarkers in cognitive impairment and dementia, especially for VCID where available biomarkers are relatively scarce.

  • Open Access

    Vascular Microbleeds Without Brain Atrophy: A Microvascular Signature of Mid-Stage 5xFAD Pathology

    by Xiuli Yang, Yuguo Li, Adnan Bibic, Jiekang Wang, Mei Wan, Wenzhen Duan, Hanzhang Lu, Zhiliang Wei

    Cerebral microbleeds are increasingly recognized as a downstream manifestation of vascular injury in Alzheimer disease (AD), arising secondary to cerebral amyloid angiopathy (CAA). Here, we examined the pathological specificity of microbleeds by comparing an amyloidosis mouse model (5xFAD) with a small-vessel disease (SVD) model characterized by vascular smooth-muscle cell loss. In vivo multimodal MRI, including gradient-echo, spin-echo, and diffusion-weighted imaging, was complemented by ex vivo high-resolution anatomical scans for validation. Both in vivo and ex vivo gradient-echo MRI consistently revealed hippocampal microbleeds in the 5xFAD model without macroscopic atrophy or ventricular enlargement, whereas no microbleeds or blood-brain barrier disruption were detected in the SVD model. Diffusion-weighted MRI further showed region-specific alterations in apparent diffusion coefficient within the midbrain of 5xFAD mice, but not in other regions or in the SVD cohort. These findings indicate that microbleeds are a pathology-specific marker of amyloid-related vascular injury rather than a general indicator of small-vessel impairment. The combined imaging evidence underscores the potential of microbleeds as a disease-specific biomarker for detecting amyloid-driven vascular fragility and refining diagnostic and therapeutic strategies for AD.


  • Open Access

    Vascular smooth muscle cell loss, but not neuroinflammation, drives cerebrovascular reactivity impairment in Alzheimer’s disease

    by Xiuli Yang, Yuguo Li, Minmin Yao, Adnan Bibic, Wenzhen Duan, Hanzhang Lu, Zhiliang Wei

    INTRODUCTION: Cerebrovascular reactivity (CVR) impairment is a key feature of Alzheimer’s disease (AD), but its mechanistic basis remains unclear. This study examined whether vascular smooth muscle cell (VSMC) loss, rather than amyloidosis or neuroinflammation, underlies CVR deficits. METHODS: Non-contrast MRI, including phase-contrast and pseudo-continuous arterial spin labeling, was performed in mouse models of amyloidosis (5xFAD), VSMC degeneration (CADASIL), and lipopolysaccharide-induced neuroinflammation. Characterization of vascular, amyloid-β, and inflammatory markers were performed for pathological assessment. RESULTS: CVR impairment emerged only when VSMC loss was present in CADASIL mice and at older ages in 5xFAD mice (9–12 months). Amyloid-β deposition occurred earlier than VSMC loss or CVR decline. Neuroinflammation primarily altered baseline cerebral blood flow without affecting CVR or VSMC integrity. DISCUSSION: These findings identify VSMC degeneration as an important driver of CVR impairment independent of cerebral amyloid angiopathy or inflammation, highlighting vascular integrity as a potential therapeutic target in AD. Highlights: Cerebrovascular reactivity (CVR) impairment occurred in 5xFAD mice only when vascular smooth muscle cell (VSMC) loss was present 5xFAD mice exhibited prominent parenchymal but minimal vascular amyloid-β deposition VSMC developmental deficiency resulted in CVR impairment in a small-vessel disease (SVD) model Neuroinflammation primarily altered baseline cerebral blood flow (CBF) without affecting CVR


  • Automatic Quality Control for Resting‐State BOLD‐Based Cerebrovascular Reactivity Mapping

    by Yutian Wang, Peiying Liu, Mingyang Li, Xirui Hou, Hanzhang Lu, Dan Wu, Zixuan Lin, Alzheimer'S Disease Neuroimaging Initiative

    Cerebrovascular reactivity (CVR) mapping based on resting‐state BOLD fMRI can be widely available for research of vascular health not only in clinical studies but also in open databases. However, as it utilizes spontaneous CO 2 fluctuations of blood as endogenous stimuli, resting‐state CVR may be prone to low SNR and reproducibility if the CO 2 fluctuation of an individual is small. The automatic identification of such poor‐quality CVR datasets is crucial for large‐scale research. Thus, in this work, we developed an automatic quality control algorithm for resting‐state CVR mapping. Utilizing a total of 51 resting‐state CVR maps acquired with three scanning protocols in each healthy participant, quality control parameters reflecting common characteristics of poor‐quality CVR, including pooled variance of different tissue types, proportion of negative voxels in gray matter, and the sensitivity of the BOLD signal to CVR, were extracted and then combined into one comprehensive quality evaluation index (QEI). We further evaluated its performance by leave‐one‐out cross‐validation and correlation analyses with test–retest reproducibility. Leave‐one‐out cross‐validation showed that QEI was significantly correlated with the reference standard of quality evaluation in all left‐out cases ( r = 0.766). Correlation analyses with test–retest reproducibility revealed significant positive correlations between the worse QEI and similarity index of CVR maps from two tests ( r = 0.809, 0.890, 0.396, and 0.654 for data from four open databases). The proposed QEI performed not only in good agreement with visual inspection but can also adapt in resting‐state CVR from multiple age groups and scanning protocols, paving the way for the clinical applications of resting‐state CVR mapping technology.

  • Open Access

    A CSF Background Suppression Scheme in Arterial Spin Labeling MRI

    by Zhiyi Hu, Wen Shi, Yifan Gou, Zihan Wang, Vivek S. Yedavalli, Doris D. Lin, Hanzhang Lu

    Arterial spin labeling (ASL) MRI suffers from low signal‐to‐noise ratio. Current background suppression (BS) methods focus on suppressing tissue signal. The present work aims to test the hypothesis that BS schemes to suppress CSF signal can produce a greater benefit, given the recent observations that water in CSF has more pulsation as part of the neurofluid circulation. We developed a CSF BS scheme that maximally suppresses residual CSF signal using two inversion pulses. Its performance was compared with regular BS and enhanced BS, both of which primarily suppress gray and white matter signals. All schemes were evaluated in single‐delay and multi‐delay pseudo‐continuous ASL (pCASL) MRI. The single‐delay scans assessed cerebral blood flow (CBF), while the multi‐delay scans measured both CBF and arterial transit time (ATT). Reproducibility was assessed using voxel‐wise coefficient of variation (CoV) and spatial Spearman correlation coefficients (Rs). When using the CSF BS scheme, CSF signals were suppressed to less than 1% of the equilibrium magnetization, with gray and white matter signals around 5% in opposite magnetization signs. Complex control/label subtraction effectively accounted for magnetization signs and obtained the correct difference signal. In single‐delay pCASL, CSF BS reduced visually apparent hyper‐ and hypo‐intensity signals in CBF maps, which were present in the tissue‐focused regular and enhanced BS schemes. These artifactual signal fluctuations were particularly pronounced in the brain‐stem regions where CSF pulsation was the most severe, but also spread along the z‐encoding direction. Quantitatively, CSF BS resulted in a lower voxel‐wise CoV (8.8%) and higher R s (0.89) in the CBF maps, compared with tissue‐focused BS (CoV 18.8%, R s 0.64). In multi‐delay pCASL, CSF BS yielded a CoV of 5.9% and 11.3% for CBF and ATT maps, respectively, outperforming both enhanced and regular BS. These results demonstrate that CSF BS can reduce spurious signals and improve image quality in ASL perfusion MRI.


  • Accelerated 3D Mapping of Whole Brain Venous Oxygenation With Nested‐Shift Under Sampling and ESPIRIT Reconstruction

    by Yiwen Hong, Dengrong Jiang, Yi Zhang, Yi‐Cheng Hsu, Hanzhang Lu, Dan Wu, Zixuan Lin

    Purpose: Non‐invasive whole brain mapping of cerebral venous oxygenation ( Y v ) is essential in brain diseases. This study aims to develop a 3D T 2 ‐Relaxation‐Under‐Velocity‐Encoding‐and‐Rapid‐Acquisition (TRU‐VERA) sequence for whole‐brain Y v quantification in both cortical and deep cerebral veins. Methods: Three Cartesian k z – t under‐sampling strategies (static, nested‐shift, and ordinary‐shift) and five reconstruction methods (SENSE, ESPIRIT, local low rank, global low rank, compressed sensing) were tested for Y v estimation accuracy. The optimized 3D TRU‐VERA was validated with TRUST and 2D TRUPC. Its reproducibility as well as sensitivity to caffeine and age‐related oxygenation changes were examined. Results: Nested‐shift under‐sampling with ESPIRIT provided optimal accuracy, achieving 1.04 × 1.04 × 2 mm 3 resolution with four eTEs in 7 min 23 s. Y v measured by 3D TRU‐VERA showed strong correlation with TRUST ( R 2 = 0.731) and TRUPC (SSS: R 2 = 0.733, deep veins: R 2 = 0.661), and had an excellent reproducibility with an intrasession CoV of 2.3% ± 1.5% for large veins and 4.0% ± 3.2% for cortical veins, and intersession CoV of 2.2% ± 1.5% and 4.0% ± 4.2%, respectively. Conclusion: The 3D TRU‐VERA sequence enables high‐resolution, reproducible, and non‐invasive assessment of venous oxygenation across the entire brain, providing vessel‐specific quantification and comprehensive evaluation of regional cerebral oxygen metabolism.


  • Open Access

    Non‐Contrast Assessment of Blood–Brain Barrier Permeability to Water: Improved Signal Modeling and Data Acquisition

    by Wen Shi, Jiani Wu, Yifan Gou, Jie Song, Zhiyi Hu, Zihan Wang, Dengrong Jiang, Rebecca Veenhuis, Leah Rubin, Yulin Ge, et al.

    Purpose: Water‐extraction‐with‐phase‐contrast‐arterial‐spin‐tagging (WEPCAST) MRI is a non‐contrast method to estimate the blood–brain barrier (BBB) permeability to water. Similar to other arterial‐spin‐labeling (ASL) based techniques, signal‐to‐noise ratio is a limitation. This study aims to enhance its reliability via theoretical and experimental improvements. Methods: We propose a generalized‐venous‐signal (GVS) model to describe the signal evolution of WEPCAST MRI, with which the control and labeled signals can be utilized to simultaneously estimate the water extraction fraction ( E ) and venous transit time (VTT). We conducted studies to test its feasibility and inter‐visit reproducibility. We further made an experimental improvement by adding a 1‐min blood T 1 scan and investigating its benefit in reducing inter‐subject variations. Results: When applying the GVS model‐based method at different locations along the superior sagittal sinus (SSS), VTT increased from anterior to posterior segments while E remained constant, consistent with known physiology. BBB permeability‐surface‐area‐product (PS) revealed a significantly lower CoV of 5.0% ± 4.1% when using the GVS method, in comparison with 8.9% ± 6.5% using the peak‐detection method ( p = 0.002). Blood T 1 was found to be 1725.7 ± 37.2 ms in males and 1799.2 ± 122.4 ms in females. After including subject‐specific blood T 1 in the parametric estimation, inter‐subject CoV in PS was found to be 6.8%, compared with a CoV of 14.2% when using an assumed blood T 1 ( p = 0.004). VTT estimated from WEPCAST was consistent with that measured with a dedicated sequence ( R = 0.757, p = 0.011). Conclusion: The reliability of WEPCAST MRI for the measurement of BBB permeability can be improved by incorporating GVS model and individual blood T 1 .