Allen Song

Overview:

The research in our lab is concerned with advancing structural and functional MRI methodologies (e.g. fast and high-resolution imaging techniques) for human brain imaging. We also aim to improve our understanding of functional brain signals, including spatiotemporal characterizations of the blood oxygenation level dependent contrast and alternative contrast mechanisms that are more directly linked to the neuronal activities. Additional effort is invested in applying and validating the developed methods to study human functional neuroanatomy.

Positions:

Professor in Radiology

Radiology
School of Medicine

Director of the Center for Brain Imaging and Analysis

Duke-UNC Center for Brain Imaging and Analysis
School of Medicine

Professor in Psychiatry and Behavioral Sciences

Psychiatry & Behavioral Sciences
School of Medicine

Professor in Neurobiology

Neurobiology
School of Medicine

Faculty Network Member of the Duke Institute for Brain Sciences

Duke Institute for Brain Sciences
Institutes and Provost's Academic Units

Member of the Duke Cancer Institute

Duke Cancer Institute
School of Medicine

Education:

Ph.D. 1995

Medical College of Wisconsin

Visiting Fellow, Laboratory Of Brain And Cognition

National Institutes of Health

Assistant Professor of Radiology, Tenure Track, Radiology

Emory University

Grants:

High Fidelity Diffusion MRI for Children with Cerebral Palsy in Stem Cell Therapy

Administered By
Duke-UNC Center for Brain Imaging and Analysis
Awarded By
National Institutes of Health
Role
Principal Investigator
Start Date
End Date

Role of cannabis on HIV-related cognitive impairment: a brain connectomics study

Administered By
Psychiatry & Behavioral Sciences, Addiction
Awarded By
National Institutes of Health
Role
Co Investigator
Start Date
End Date

Neuroimaging of Visual Attention in Aging

Administered By
Psychiatry & Behavioral Sciences, Geriatric Behavioral Health
Awarded By
National Institutes of Health
Role
Co Investigator
Start Date
End Date

MRI data fusion to investigate effects of drug abuse on HIV neurological complications

Administered By
Psychiatry & Behavioral Sciences, Addiction
Awarded By
National Institutes of Health
Role
Co Investigator
Start Date
End Date

NCANDA Research Project Site: Duke

Administered By
Psychiatry, Child & Family Mental Health & Developmental Neuroscience
Awarded By
National Institutes of Health
Role
Collaborator
Start Date
End Date

Publications:

Plasma cytokines associated with febrile status epilepticus in children: A potential biomarker for acute hippocampal injury.

OBJECTIVE: Our aim was to explore the association between plasma cytokines and febrile status epilepticus (FSE) in children, as well as their potential as biomarkers of acute hippocampal injury. METHODS: Analysis was performed on residual samples of children with FSE (n = 33) as part of the Consequences of Prolonged Febrile Seizures in Childhood study (FEBSTAT) and compared to children with fever (n = 17). Magnetic resonance imaging (MRI) was obtained as part of FEBSTAT within 72 h of FSE. Cytokine levels and ratios of antiinflammatory versus proinflammatory cytokines in children with and without hippocampal T2 hyperintensity were assessed as biomarkers of acute hippocampal injury after FSE. RESULTS: Levels of interleukin (IL)-8 and epidermal growth factor (EGF) were significantly elevated after FSE in comparison to controls. IL-1β levels trended higher and IL-1RA trended lower following FSE, but did not reach statistical significance. Children with FSE were found to have significantly lower ratios of IL-1RA/IL-1β and IL-1RA/IL-8. Specific levels of any one individual cytokine were not associated with FSE. However, lower ratios of IL-1RA/IL-1β, IL-1RA/1L-6, and IL-1RA/ IL-8 were all associated with FSE. IL-6 and IL-8 levels were significantly higher and ratios of IL-1RA/IL-6 and IL-1RA/IL-8 were significantly lower in children with T2 hippocampal hyperintensity on MRI after FSE in comparison to those without hippocampal signal abnormalities. Neither individual cytokine levels nor ratios of IL-1RA/IL-1β or IL-1RA/IL-8 were predictive of MRI changes. However, a lower ratio of IL-1RA/IL-6 was strongly predictive (odds ratio [OR] 21.5, 95% confidence interval [CI] 1.17-393) of hippocampal T2 hyperintensity after FSE. SIGNIFICANCE: Our data support involvement of the IL-1 cytokine system, IL-6, and IL-8 in FSE in children. The identification of the IL-1RA/IL-6 ratio as a potential biomarker of acute hippocampal injury following FSE is the most significant finding. If replicated in another study, the IL-1RA/IL-6 ratio could represent a serologic biomarker that offers rapid identification of patients at risk for ultimately developing mesial temporal lobe epilepsy (MTLE).
Authors
Gallentine, WB; Shinnar, S; Hesdorffer, DC; Epstein, L; Nordli, DR; Lewis, DV; Frank, LM; Seinfeld, S; Shinnar, RC; Cornett, K; Liu, B; Moshé, SL; Sun, S; FEBSTAT Investigator Team,
MLA Citation
Gallentine, William B., et al. “Plasma cytokines associated with febrile status epilepticus in children: A potential biomarker for acute hippocampal injury.Epilepsia, vol. 58, no. 6, June 2017, pp. 1102–11. Pubmed, doi:10.1111/epi.13750.
URI
https://scholars.duke.edu/individual/pub1250545
PMID
28448686
Source
pubmed
Published In
Epilepsia
Volume
58
Published Date
Start Page
1102
End Page
1111
DOI
10.1111/epi.13750

Differential developmental trajectories of magnetic susceptibility in human brain gray and white matter over the lifespan.

As indicated by several recent studies, magnetic susceptibility of the brain is influenced mainly by myelin in the white matter and by iron deposits in the deep nuclei. Myelination and iron deposition in the brain evolve both spatially and temporally. This evolution reflects an important characteristic of normal brain development and ageing. In this study, we assessed the changes of regional susceptibility in the human brain in vivo by examining the developmental and ageing process from 1 to 83 years of age. The evolution of magnetic susceptibility over this lifespan was found to display differential trajectories between the gray and the white matter. In both cortical and subcortical white matter, an initial decrease followed by a subsequent increase in magnetic susceptibility was observed, which could be fitted by a Poisson curve. In the gray matter, including the cortical gray matter and the iron-rich deep nuclei, magnetic susceptibility displayed a monotonic increase that can be described by an exponential growth. The rate of change varied according to functional and anatomical regions of the brain. For the brain nuclei, the age-related changes of susceptibility were in good agreement with the findings from R2* measurement. Our results suggest that magnetic susceptibility may provide valuable information regarding the spatial and temporal patterns of brain myelination and iron deposition during brain maturation and ageing.
Authors
Li, W; Wu, B; Batrachenko, A; Bancroft-Wu, V; Morey, RA; Shashi, V; Langkammer, C; De Bellis, MD; Ropele, S; Song, AW; Liu, C
MLA Citation
Li, Wei, et al. “Differential developmental trajectories of magnetic susceptibility in human brain gray and white matter over the lifespan..” Hum Brain Mapp, vol. 35, no. 6, June 2014, pp. 2698–713. Pubmed, doi:10.1002/hbm.22360.
URI
https://scholars.duke.edu/individual/pub962825
PMID
24038837
Source
pubmed
Published In
Hum Brain Mapp
Volume
35
Published Date
Start Page
2698
End Page
2713
DOI
10.1002/hbm.22360

Application of k-space energy spectrum analysis for inherent and dynamic B0 mapping and deblurring in spiral imaging.

Spiral imaging is vulnerable to spatial and temporal variations of the amplitude of the static magnetic field (B(0)) caused by susceptibility effects, eddy currents, chemical shifts, subject motion, physiological noise, and system instabilities, resulting in image blurring. Here, a novel off-resonance correction method is proposed to address these issues. A k-space energy spectrum analysis algorithm is first applied to inherently and dynamically generate a B(0) map from the k-space data at each time point, without requiring any additional data acquisition, pulse sequence modification, or phase unwrapping. A simulated phase evolution rewinding algorithm and an automatic residual deblurring algorithm are then used to correct for the blurring caused by both spatial and temporal B(0) variations, resulting in a high spatial and temporal fidelity. This method is validated against conventional B(0) mapping and deblurring methods, and its advantages for dynamic MRI applications are demonstrated in functional MRI studies.
Authors
Truong, T-K; Chen, N-K; Song, AW
MLA Citation
Truong, Trong-Kha, et al. “Application of k-space energy spectrum analysis for inherent and dynamic B0 mapping and deblurring in spiral imaging..” Magn Reson Med, vol. 64, no. 4, Oct. 2010, pp. 1121–27. Pubmed, doi:10.1002/mrm.22485.
URI
https://scholars.duke.edu/individual/pub736834
PMID
20564589
Source
pubmed
Published In
Magn Reson Med
Volume
64
Published Date
Start Page
1121
End Page
1127
DOI
10.1002/mrm.22485

Single-shot dual-z-shimmed sensitivity-encoded spiral-in/out imaging for functional MRI with reduced susceptibility artifacts.

Blood oxygenation level-dependent (BOLD) functional MRI (fMRI) can be severely hampered by signal loss due to susceptibility-induced static magnetic field (B(0)) inhomogeneities near air/tissue interfaces. A single-shot spiral-in/out sequence with a z-shim gradient embedded between the two acquisitions was previously proposed to efficiently recover the signal. However, despite promising results, this technique had several limitations, which are addressed here as follows. First, by adding a second z-shim gradient before the spiral-in acquisition and optimizing both z-shim gradients slice-by-slice, a significantly more uniform signal recovery can be achieved. Second, by acquiring a B(0) map, the optimal z-shim gradients can be directly, efficiently, and accurately determined for each subject. Third, by complementing the z-shimming approach with sensitivity encoding (SENSE), the in-plane spatial resolution can be increased and, hence, susceptibility artifacts further reduced, while maintaining a high temporal resolution for fMRI applications. These advantages are demonstrated in human functional studies.
Authors
Truong, T-K; Song, AW
MLA Citation
Truong, Trong-Kha, and Allen W. Song. “Single-shot dual-z-shimmed sensitivity-encoded spiral-in/out imaging for functional MRI with reduced susceptibility artifacts..” Magn Reson Med, vol. 59, no. 1, Jan. 2008, pp. 221–27. Pubmed, doi:10.1002/mrm.21473.
URI
https://scholars.duke.edu/individual/pub736841
PMID
18050341
Source
pubmed
Published In
Magnetic Resonance in Medicine
Volume
59
Published Date
Start Page
221
End Page
227
DOI
10.1002/mrm.21473

The spatial and temporal characteristics of the apparent-diffusion-coefficient-dependent fMRI signal changes during visual stimulation.

The blood oxygenation level dependent (BOLD) contrast has been commonly used to detect fMRI signal. The majority of the BOLD signals are believed to arise from the venous and capillary networks. However, only those from the capillaries are spatially close to the neuronal activities, while the signals from large veins could be distant, rendering the overall localization inaccurate. In recent years, an alternative contrast using arterial spin labeled (ASL) perfusion imaging techniques has been proposed for predominant capillary sensitivity. Such acquisition methods, however, are intrinsically limited in temporal resolution and spatial coverage. Another contrast mechanism, free of such constraints, is based on the apparent diffusion coefficient (ADC) changes during brain activation using isotropic diffusion weighting. It has been shown that these changes are synchronized with brain activation and that they, as a whole, temporally precede BOLD activation, suggesting significant upstream arterial contribution. Moreover, the spatial overlaps between the upstream ADC and downstream BOLD activations are shown to be more localized in the capillaries, which are the temporal and spatial middle ground. In this paper, we sought to further investigate the temporal and spatial characteristics of ADC contrast with additional arterial signal suppression. Also, a pixel-based evaluation was performed in conjunction with the averaged global assessment. It was found that in addition to the known spatial discrepancy and global timing advance compared to the BOLD signal, the ADC activation endured significant temporal heterogeneities. Such fine spatial and temporal assessment could help characterize the exact signal sources of ADC contrast, and ultimately achieve exclusive capillary sensitivity.
Authors
Song, AW; Gangstead, SL
MLA Citation
Song, Allen W., and Stacey L. Gangstead. “The spatial and temporal characteristics of the apparent-diffusion-coefficient-dependent fMRI signal changes during visual stimulation..” J Neural Eng, vol. 1, no. 1, Mar. 2004, pp. 32–38. Pubmed, doi:10.1088/1741-2560/1/1/005.
URI
https://scholars.duke.edu/individual/pub686701
PMID
15876620
Source
pubmed
Published In
Journal of Neural Engineering
Volume
1
Published Date
Start Page
32
End Page
38
DOI
10.1088/1741-2560/1/1/005