Frontiers

User Collection Public
UNC logo

UNC-authored articles published by Frontiers

Works (762)

View results as:
List List Masonry Masonry Slideshow Slideshow
Sort the listing of items  

541. Whole genome sequencing of Enterovirus species C isolates by high-throughput sequencing: Development of generic primers

Title Tesim:
Whole genome sequencing of Enterovirus species C isolates by high-throughput sequencing: Development of generic primers
Creator:
Volle, R., Blondel, B., Bessaud, M., Delpeyroux, F., Njouom, R., Razafindratsimandresy, R., Polston, P., Rakoto-Andrianarivelo, M., Sadeuh-Mba, S.A., and Joffret, M.-L.
Date of publication:
2016
Abstract Tesim:
Enteroviruses are among the most common viruses infecting humans and can cause diverse clinical syndromes ranging from minor febrile illness to severe and potentially fatal diseases. Enterovirus species C (EV-C) consists of more than 20 types, among which the three serotypes of polioviruses, the etiological agents of poliomyelitis, are included. Biodiversity and evolution of EV-C genomes are shaped by frequent recombination events. Therefore, identification and characterization of circulating EV-C strains require the sequencing of different genomic regions. A simple method was developed to quickly sequence the entire genome of EV-C isolates. Four overlapping fragments were produced separately by RT-PCR performed with generic primers. The four amplicons were then pooled and purified prior to being sequenced by a high-throughput technique. The method was assessed on a panel of EV-Cs belonging to a wide-range of types. It can be used to determine full-length genome sequences through de novo assembly of thousands of reads. It was also able to discriminate reads from closely related viruses in mixtures. By decreasing the workload compared to classical Sanger-based techniques, this method will serve as a precious tool for sequencing large panels of EV-Cs isolated in cell cultures during environmental surveillance or from patients, including vaccine-derived polioviruses.
Resource type:
Article
Affiliation Label Tesim:
Department of Health Behavior
DOI:
https://doi.org/10.17615/nybg-g321
Edition:
Publisher
ISSN:
1664-302X
Journal Issue:
AUG
Journal Title:
Frontiers in Microbiology
Journal Volume:
7
Keyword:
Poliovirus, De novo assembly, Recombination, Enterovirus species C, and High-throughput sequencing
Language Label:
English
License Label:
Attribution 3.0 United States
Other Affiliation:
Institut Pasteur, Centre Pasteur du Cameroun, Institut Pasteur de Madagascar, and Université Ankatso
Person:
Volle, R., Blondel, B., Bessaud, M., Delpeyroux, F., Njouom, R., Razafindratsimandresy, R., Polston, P., Rakoto-Andrianarivelo, M., Sadeuh-Mba, S.A., and Joffret, M.-L.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
patsy_polston_2

542. Super-resolution imaging of C-type lectin spatial rearrangement within the dendritic cell plasma membrane at fungal microbe contact sites

Title Tesim:
Super-resolution imaging of C-type lectin spatial rearrangement within the dendritic cell plasma membrane at fungal microbe contact sites
Creator:
Liu, P., Lidke, K.A., Wester, M.J., Graus, M.S., Itano, M.S., Neumann, A.K., Thompson, N.L., Pehlke, C., and Jacobson, K.
Date of publication:
2014
Abstract Tesim:
Dendritic cells express DC-SIGN and CD206, C-type lectins (CTLs) that bind a variety of pathogens and may facilitate pathogen uptake for subsequent antigen presentation. Both proteins form punctate membrane nanodomains (~80 nm) on naïve cells. We analyzed the spatiotemporal distribution of CTLs following host-fungal particle contact using confocal microscopy and three distinct methods of cluster identification and measurement of receptor clusters in super-resolution datasets: DBSCAN, Pair Correlation and a custom implementation of the Getis spatial statistic. Quantitative analysis of confocal and super-resolution images demonstrated that CTL nanodomains become concentrated in the contact site relative to non-contact membrane after the first hour of exposure and established that this recruitment is sustained out to 4 h. DC-SIGN nanodomains in fungal contact sites exhibit a 70% area increase and a 38% decrease in interdomain separation. Contact site CD206 nanodomains possess 90% greater area and 42% lower interdomain separation relative to non-contact regions. Contact site CTL clusters appear as disk-shaped domains of approximately 150–175 nm in diameter. The increase in length scale of CTL nanostructure in contact sites suggests that the smaller nanodomains on resting membranes may merge during fungal recognition, or that they become packed closely enough to achieve sub-resolution inter-domain edge separations of <30 nm. This study provides evidence of local receptor spatial rearrangements on the nanoscale that occur in the plasma membrane upon pathogen binding and may direct important signaling interactions required to recognize and respond to the presence of a relatively large pathogen.
Resource type:
Article
Affiliation Label Tesim:
Department of Cell Biology and Physiology and Department of Chemistry
DOI:
https://doi.org/10.17615/d3ar-2q05
Identifier:
https://dx.doi.org/10.3389/fphy.2014.00046
ISSN:
2296-424X
Journal Title:
Frontiers in Physics
Journal Volume:
2
Keyword:
CD206, Zymosan, Mannan, DC-SIGN, Dendritic cell, and dSTORM
Language Label:
English
License Label:
Attribution 3.0 United States
Other Affiliation:
University of New Mexico
Page End:
17
Page Start:
1
Person:
Liu, P., Lidke, K.A., Wester, M.J., Graus, M.S., Itano, M.S., Neumann, A.K., Thompson, N.L., Pehlke, C., and Jacobson, K.
Rights Statement Label:
In Copyright
Source:
michelle_itano_4

543. Expression of the pluripotency transcription factor OCT4 in the normal and aberrant mammary gland

Title Tesim:
Expression of the pluripotency transcription factor OCT4 in the normal and aberrant mammary gland
Creator:
Filgueira, L., Blancafort, P., Beltran, A.S., Mathews, M.M., Hepworth, A.R., Hassiotou, F., Stuebe, A.M., and Hartmann, P.E.
Date of publication:
2013
Abstract Tesim:
Breast cancers with lactating features, some of which are associated with pregnancy and lactation, are often poorly differentiated, lack estrogen receptor, progesterone receptor, and HER2 expression and have high mortality. Very little is known about the molecular mechanisms that drive uncontrolled cell proliferation in these tumors and confer lactating features. We have recently reported expression of OCT4 and associated embryonic stem cell self-renewal genes in the normal lactating breast and breastmilk stem cells (hBSCs). This prompted us to examine OCT4 expression in breast cancers with lactating features and compare it with that observed during normal lactation, using rare specimens of human lactating breast. In accordance with previous literature, the normal resting breast (from non-pregnant, non-lactating women) showed minimal OCT4 nuclear expression (0.9%). However, this increased in the normal lactating breast (11.4%), with further increase in lactating adenomas, lactating carcinomas, and pregnancy-associated breast cancer (30.7-48.3%). OCT4 was expressed in the epithelium and at lower levels in the stroma, and was co-localized with NANOG. Comparison of normal non-tumorigenic hBSCs with OCT4-overexpressing tumorigenic breast cell lines (OTBCs) demonstrated upregulation of OCT4, SOX2, and NANOG in both systems, but OTBCs expressed OCT4 at significantly higher levels than SOX2 and NANOG. Similar to hBSCs, OTBCs displayed multi-lineage differentiation potential, including the ability to differentiate into functional lactocytes synthesizing milk proteins both in vitro and in vivo. Based on these findings, we propose a hypothesis of normal and malignant transformation in the breast, which centers on OCT4 and its associated gene network. Although minimal expression of these embryonic genes can be seen in the breast in its resting state throughout life, a controlled program of upregulation of this gene network may be a potential regulator of the normal remodeling of the breast toward a milk-secretory organ during pregnancy and lactation. Deregulation of this gene network either within or outside pregnancy and lactation may lead to aberrant breast cell proliferation and malignant transformation, suggesting a role of these genes in both normal lactation and breast oncogenesis.
Resource type:
Article
Affiliation Label Tesim:
Department of Pharmacology, Translational Pathology Laboratory, and Department of Obstetrics and Gynecology
DOI:
https://doi.org/10.17615/3ccm-vz55
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fonc.2013.00079
ISSN:
2234-943X
Journal Title:
Frontiers in Oncology
Journal Volume:
3-Apr
Keyword:
Self-renewal, Cancer stem cell, Adult stem cell, Breast cancer, Transcription factors, Mammary gland, OCT4, and Breastmilk
Language Label:
English
License Label:
Attribution 3.0 United States
Other Affiliation:
The University of Western Australia
Person:
Filgueira, L., Blancafort, P., Beltran, A.S., Mathews, M.M., Hepworth, A.R., Hassiotou, F., Stuebe, A.M., and Hartmann, P.E.
Rights Statement Label:
In Copyright
Source:
adriana_beltran_5

544. Synergy of image analysis for animal and human neuroimaging supports translational research on drug abuse

Title Tesim:
Synergy of image analysis for animal and human neuroimaging supports translational research on drug abuse
Creator:
Grewen, K., Oguz, I., Lee, J., An, H., Lin, W., Johns, J., McMurray, M., Styner, M.A., Gouttard, S., and Gerig, G.
Date of publication:
2011
Abstract Tesim:
The use of structural magnetic resonance imaging (sMRI) and diffusion tensor imaging (DTI) in animal models of neuropathology is of increasing interest to the neuroscience community. In this work, we present our approach to create optimal translational studies that include both animal and human neuroimaging data within the frameworks of a study of post-natal neuro-development in intra-uterine cocaine-exposure. We propose the use of non-invasive neuroimaging to study developmental brain structural and white matter pathway abnormalities via sMRI and DTI, as advanced MR imaging technology is readily available and automated image analysis methodology have recently been transferred from the human to animal imaging setting. For this purpose, we developed a synergistic, parallel approach to imaging and image analysis for the human and the rodent branch of our study. We propose an equivalent design in both the selection of the developmental assessment stage and the neuroimaging setup. This approach brings significant advantages to study neurobiological features of early brain development that are common to animals and humans but also preserve analysis capabilities only possible in animal research. This paper presents the main framework and individual methods for the proposed cross-species study design, as well as preliminary DTI cross-species comparative results in the intra-uterine cocaine-exposure study.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry, Department of Computer Science, Biomedical Research Imaging Center, and Department of Psychology
DOI:
https://doi.org/10.17615/m4qr-sg96
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fpsyt.2011.00053
ISSN:
1664-0640
Journal Issue:
OCT
Journal Title:
Frontiers in Psychiatry
Journal Volume:
2
Keyword:
Drug abuse, Small animal imaging, White matter pathways, Neonatal neuroimaging, Brain segmentation, Magnetic resonance imaging, and Diffusion tensor imaging
Other Affiliation:
University of Utah
Person:
Grewen, K., Oguz, I., Lee, J., An, H., Lin, W., Johns, J., McMurray, M., Styner, M.A., Gouttard, S., and Gerig, G.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
martin_styner_94

545. Use of high resolution 3D diffusion tensor imaging to study brain white matter development in live neonatal rats

Title Tesim:
Use of high resolution 3D diffusion tensor imaging to study brain white matter development in live neonatal rats
Creator:
An, H., Oguz, I., Johns, J.M., Lin, W., Styner, M.A., McMurray, M.S., Cai, Y., and Yuan, H.
Date of publication:
2011
Abstract Tesim:
High resolution diffusion tensor imaging (DTI) can provide important information on brain development, yet it is challenging in live neonatal rats due to the small size of neonatal brain and motion-sensitive nature of DTI. Imaging in live neonatal rats has clear advantages over fixed brain scans, as longitudinal and functional studies would be feasible to understand neuro-developmental abnormalities. In this study, we developed imaging strategies that can be used to obtain high resolution 3D DTI images in live neonatal rats at postnatal day 5 (PND5) and PND14, using only 3 h of imaging acquisition time. An optimized 3D DTI pulse sequence and appropriate animal setup to minimize physiological motion artifacts are the keys to successful high resolution 3D DTI imaging. Thus, a 3D rapid acquisition relaxation enhancement DTI sequence with twin navigator echoes was implemented to accelerate imaging acquisition time and minimize motion artifacts. It has been suggested that neonatal mammals possess a unique ability to tolerate mild-to-moderate hypothermia and hypoxia without long term impact. Thus, we additionally utilized this ability to minimize motion artifacts in magnetic resonance images by carefully suppressing the respiratory rate to around 15/min for PND5 and 30/min for PND14 using mild-to-moderate hypothermia. These imaging strategies have been successfully implemented to study how the effect of cocaine exposure in dams might affect brain development in their rat pups. Image quality resulting from this in vivo DTI study was comparable to ex vivo scans. fractional anisotropy values were also similar between the live and fixed brain scans. The capability of acquiring high quality in vivo DTI imaging offers a valuable opportunity to study many neurological disorders in brain development in an authentic living environment.
Resource type:
Article
Affiliation Label Tesim:
Department of Radiology and Department of Psychology and Neuroscience
DOI:
https://doi.org/10.17615/t0fk-ss70
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fpsyt.2011.00054
ISSN:
1664-0640
Journal Issue:
OCT
Journal Title:
Frontiers in Psychiatry
Journal Volume:
2
Keyword:
Brain development, Diffusion tensor imaging, White matter, Magnetic resonance imaging, and Neonatal rats
Person:
An, H., Oguz, I., Johns, J.M., Lin, W., Styner, M.A., McMurray, M.S., Cai, Y., and Yuan, H.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
martin_styner_93

546. Hippocampal shape and volume changes with antipsychotics in early stage psychotic illness

Title Tesim:
Hippocampal shape and volume changes with antipsychotics in early stage psychotic illness
Creator:
Wang, L., Barch, D., Lieberman, J.A., Styner, M., Harms, M.P., and Mamah, D.
Date of publication:
2012
Abstract Tesim:
Progression of hippocampal shape and volume abnormalities has been described in psychotic disorders such as schizophrenia. However it is unclear how specific antipsychotic medications influence the development of hippocampal structure.We conducted a longitudinal, randomized, controlled, multisite, double-blind study involving 14 academic medical centers (United States 11, Canada 1, Netherlands 1, and England 1). One hundred thirty-four first-episode psychosis patients (receiving either haloperidol [HAL] or olanzapine [OLZ]) and 51 healthy controls were followed for up to 104 weeks using magnetic resonance imaging and large-deformation high-dimensional brain mapping of the hippocampus. Changes in hippocampal volume and shape metrics (i.e., percentage of negative surface vertex slopes, and surface deformation) were evaluated. Mixed-models analysis did not show a significant group-by-time interaction for hippocampal volume. However, the cumulative distribution function of hippocampal surface vertex slopes showed a notable left shift with HAL treatment compared to OLZ treatment and tocontrols. OLZ treatment was associated with a significantly lower percentage of "large magnitude" negative surface vertex slopes compared to HAL treatment (p = 0.004). Surface deformation maps however did not localize any hippocampal regions that differentially contracted over time with OLZ treatment, after FDR correction. These results indicate that surface analysis provides supplementary information to volumetry in detecting differential treatment effects of the hippocampus. Our results suggest that OLZ is associated with less longitudinal hippocampal surface deformation than HAL, however the hippocampal regions affected appear to be variable across patients.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry
DOI:
https://doi.org/10.17615/03d2-tj06
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fpsyt.2012.00096
ISSN:
1664-0640
Journal Issue:
NOV
Journal Title:
Frontiers in Psychiatry
Journal Volume:
3
Keyword:
Olanzapine, Schizophrenia, Psychosis, Haloperidol, MRI, and Hippocampus
License Label:
Attribution 3.0 United States
Other Affiliation:
Northwestern University Feinberg School of Medicine, Washington University, and Columbia University Medical Center
Person:
Wang, L., Barch, D., Lieberman, J.A., Styner, M., Harms, M.P., and Mamah, D.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
martin_styner_90

547. The UNC-Wisconsin rhesus macaque neurodevelopment database: A structural MRI and DTI database of early postnatal development

Title Tesim:
The UNC-Wisconsin rhesus macaque neurodevelopment database: A structural MRI and DTI database of early postnatal development
Creator:
Knickmeyer, R.C., Grauer, M., Davis, B., Lubach, G.R., Styner, M.A., Young, J.T., Shi, Y., Niethammer, M., Coe, C.L., Alexander, A.L., and Budin, F.
Date of publication:
2017
Abstract Tesim:
Rhesus macaques are commonly used as a translational animal model in neuroimaging and neurodevelopmental research. In this report, we present longitudinal data from both structural and diffusion MRI images generated on a cohort of 34 typically developing monkeys from 2 weeks to 36 months of age. All images have been manually skull stripped and are being made freely available via an online repository for use by the research community.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry and Department of Computer Science
DOI:
https://doi.org/10.17615/t086-1w19
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fnins.2017.00029
ISSN:
1662-4548
Journal Issue:
FEB
Journal Title:
Frontiers in Neuroscience
Journal Volume:
11
Keyword:
Magnetic resonance imaging, Brain development, Non-human primate, Diffusion tensor imaging, Computational atlases, Neuroimaging, and Macaque
License Label:
Attribution 3.0 United States
Other Affiliation:
Kitware Inc. and University of Wisconsin-Madison
Person:
Knickmeyer, R.C., Grauer, M., Davis, B., Lubach, G.R., Styner, M.A., Young, J.T., Shi, Y., Niethammer, M., Coe, C.L., Alexander, A.L., and Budin, F.
Publisher:
Frontiers Research Foundation
Rights Statement Label:
In Copyright
Source:
martin_styner_71

548. Neonatal White Matter Maturation Is Associated With Infant Language Development

Title Tesim:
Neonatal White Matter Maturation Is Associated With Infant Language Development
Creator:
Wadhwa, P.D., Styner, M., Rasmussen, J.M., Gilmore, J.H., Entringer, S., Buss, C., Overfeld, J., and Sket, G.M.
Date of publication:
2019
Abstract Tesim:
Background: While neonates have no sophisticated language skills, the neural basis for acquiring this function is assumed to already be present at birth. Receptive language is measurable by 6 months of age and meaningful speech production by 10–18 months of age. Fiber tracts supporting language processing include the corpus callosum (CC), which plays a key role in the hemispheric lateralization of language; the left arcuate fasciculus (AF), which is associated with syntactic processing; and the right AF, which plays a role in prosody and semantics. We examined if neonatal maturation of these fiber tracts is associated with receptive language development at 12 months of age. Methods: Diffusion-weighted imaging (DWI) was performed in 86 infants at 26.6 ± 12.2 days post-birth. Receptive language was assessed via the MacArthur-Bates Communicative Development Inventory at 12 months of age. Tract-based fractional anisotropy (FA) was determined using the NA-MIC atlas-based fiber analysis toolkit. Associations between neonatal regional FA, adjusted for gestational age at birth and age at scan, and language development at 12 months of age were tested using ANOVA models. Results: After multiple comparisons correction, higher neonatal FA was positively associated with receptive language at 12 months of age within the genu (p < 0.001), rostrum (p < 0.001), and tapetum (p < 0.001) of the CC and the left fronto-parietal AF (p = 0.008). No significant clusters were found in the right AF. Conclusion: Microstructural development of the CC and the AF in the newborn is associated with receptive language at 12 months of age, demonstrating that interindividual variation in white matter microstructure is relevant for later language development, and indicating that the neural foundation for language processing is laid well ahead of the majority of language acquisition. This suggests that some origins of impaired language development may lie in the intrauterine and potentially neonatal period of life. Understanding how interindividual differences in neonatal brain maturity relate to the acquisition of function, particularly during early development when the brain is in an unparalleled window of plasticity, is key to identifying opportunities for harnessing neuroplasticity in health and disease.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry
DOI:
https://doi.org/10.17615/p2sf-mf59
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fnhum.2019.00434
ISSN:
1662-5161
Journal Title:
Frontiers in Human Neuroscience
Journal Volume:
13
Keyword:
infant language development, receptive language development, white matter development, neonatal neuroimaging, and diffusion tensor imaging
License Label:
Attribution 3.0 United States
ORCID:
Other Affiliation:
University of California, Irvine, , and Berlin Institute of Health
Person:
Wadhwa, P.D., Styner, M., Rasmussen, J.M., Gilmore, J.H., Entringer, S., Buss, C., Overfeld, J., and Sket, G.M.
Publisher:
Frontiers Media S.A.
Source:
martin_styner_29

549. A Novel Method for High-Dimensional Anatomical Mapping of Extra-Axial Cerebrospinal Fluid: Application to the Infant Brain

Title Tesim:
A Novel Method for High-Dimensional Anatomical Mapping of Extra-Axial Cerebrospinal Fluid: Application to the Infant Brain
Creator:
Girault, J.B., Schultz, R.T., Estes, A.M., Evans, A.C., Botteron, K.N., Gerig, G., Hazlett, H.C., Kim, S.H., Styner, M.A., Mostapha, M., Dager, S.R., Pizer, S.M., McKinstry, R.C., Piven, J., and Shen, M.D.
Date of publication:
2020
Abstract Tesim:
Cerebrospinal fluid (CSF) plays an essential role in early postnatal brain development. Extra-axial CSF (EA-CSF) volume, which is characterized by CSF in the subarachnoid space surrounding the brain, is a promising marker in the early detection of young children at risk for neurodevelopmental disorders. Previous studies have focused on global EA-CSF volume across the entire dorsal extent of the brain, and not regionally-specific EA-CSF measurements, because no tools were previously available for extracting local EA-CSF measures suitable for localized cortical surface analysis. In this paper, we propose a novel framework for the localized, cortical surface-based analysis of EA-CSF. The proposed processing framework combines probabilistic brain tissue segmentation, cortical surface reconstruction, and streamline-based local EA-CSF quantification. The quantitative analysis of local EA-CSF was applied to a dataset of typically developing infants with longitudinal MRI scans from 6 to 24 months of age. There was a high degree of consistency in the spatial patterns of local EA-CSF across age using the proposed methods. Statistical analysis of local EA-CSF revealed several novel findings: several regions of the cerebral cortex showed reductions in EA-CSF from 6 to 24 months of age, and specific regions showed higher local EA-CSF in males compared to females. These age-, sex-, and anatomically-specific patterns of local EA-CSF would not have been observed if only a global EA-CSF measure were utilized. The proposed methods are integrated into a freely available, open-source, cross-platform, user-friendly software tool, allowing neuroimaging labs to quantify local extra-axial CSF in their neuroimaging studies to investigate its role in typical and atypical brain development.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry and Department of Computer Science
DOI:
https://doi.org/10.17615/7b4r-ae21
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fnins.2020.561556
ISSN:
1662-4548
Journal Title:
Frontiers in Neuroscience
Journal Volume:
14
Keyword:
neurodevelopmental disorders, structural MRI, brain development, surface analysis, Laplacian PDE, extra-axial cerebrospinal fluid, autism, and EA-CSF
License Label:
Attribution 3.0 United States
Other Affiliation:
University of Pennsylvania, University of Washington, McGill University, Washington University School of Medicine, and New York University
Person:
Girault, J.B., Schultz, R.T., Estes, A.M., Evans, A.C., Botteron, K.N., Gerig, G., Hazlett, H.C., Kim, S.H., Styner, M.A., Mostapha, M., Dager, S.R., Pizer, S.M., McKinstry, R.C., Piven, J., and Shen, M.D.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
martin_styner_14

550. A 3D Fully Convolutional Neural Network With Top-Down Attention-Guided Refinement for Accurate and Robust Automatic Segmentation of Amygdala and Its Subnuclei

Title Tesim:
A 3D Fully Convolutional Neural Network With Top-Down Attention-Guided Refinement for Accurate and Robust Automatic Segmentation of Amygdala and Its Subnuclei
Creator:
Liu, Y., Nacewicz, B.M., Ferrazzano, P.A., Styner, M.A., Alexander, A.L., Kirk, G.R., Adluru, N., and Zhao, G.
Date of publication:
2020
Abstract Tesim:
Recent advances in deep learning have improved the segmentation accuracy of subcortical brain structures, which would be useful in neuroimaging studies of many neurological disorders. However, most existing deep learning based approaches in neuroimaging do not investigate the specific difficulties that exist in segmenting extremely small but important brain regions such as the subnuclei of the amygdala. To tackle this challenging task, we developed a dual-branch dilated residual 3D fully convolutional network with parallel convolutions to extract more global context and alleviate the class imbalance issue by maintaining a small receptive field that is just the size of the regions of interest (ROIs). We also conduct multi-scale feature fusion in both parallel and series to compensate the potential information loss during convolutions, which has been shown to be important for small objects. The serial feature fusion enabled by residual connections is further enhanced by a proposed top-down attention-guided refinement unit, where the high-resolution low-level spatial details are selectively integrated to complement the high-level but coarse semantic information, enriching the final feature representations. As a result, the segmentations resulting from our method are more accurate both volumetrically and morphologically, compared with other deep learning based approaches. To the best of our knowledge, this work is the first deep learning-based approach that targets the subregions of the amygdala. We also demonstrated the feasibility of using a cycle-consistent generative adversarial network (CycleGAN) to harmonize multi-site MRI data, and show that our method generalizes well to challenging traumatic brain injury (TBI) datasets collected from multiple centers. This appears to be a promising strategy for image segmentation for multiple site studies and increased morphological variability from significant brain pathology.
Resource type:
Article
Affiliation Label Tesim:
Department of Psychiatry
DOI:
https://doi.org/10.17615/vznx-k425
Edition:
Publisher
Identifier:
https://dx.doi.org/10.3389/fnins.2020.00260
ISSN:
1662-4548
Journal Title:
Frontiers in Neuroscience
Journal Volume:
14
Keyword:
segmentation, amygdala, deep learning, fully convolutional neural network, generalization, harmonization, and structural MRI
License Label:
Attribution 3.0 United States
Other Affiliation:
University of Wisconsin-Madison
Person:
Liu, Y., Nacewicz, B.M., Ferrazzano, P.A., Styner, M.A., Alexander, A.L., Kirk, G.R., Adluru, N., and Zhao, G.
Publisher:
Frontiers Media S.A.
Rights Statement Label:
In Copyright
Source:
martin_styner_13