Neil Gelman

White matter abnormalities in autism detected through transverse relaxation time imaging.

While neuroimaging studies have reported neurobiological abnormalities in autism, the underlying tissue abnormalities remain unclear. Quantitative transverse relaxation time (T2) imaging permits the examination of tissue abnormalities in vivo, with increased T2 largely reflecting increased tissue water. Blood flow and the presence of tissue iron may also affect T2. In this study, we used voxel-based relaxometry of the cerebrum and global averages to examine T2 abnormalities in autism. Nineteen males with autism (age: 9.2 +/- 3.0 years) and 20 male controls (age: 10.7 +/- 2.9 years) underwent magnetic resonance imaging at 3.0 T. Quantitative T2 maps, generated through gradient echo sampling of the free induction decay and echo, were segmented into gray matter, white matter, and cerebrospinal fluid. Average cerebral gray and white matter T2 were determined and compared between groups. To assess localized T2 differences, the quantitative T2 maps were warped to a template created for this study, smoothed, and compared using statistical parametric mapping. Patients with autism had an increase in average cerebral white matter T2, although no group differences were seen in average cerebral gray matter T2. Patients with autism also had bilateral regional T2 increases in the gray matter and associated white matter of the parietal lobes (primary sensory association areas) and occipital lobes (visual association areas) and in the white matter within the supplementary motor areas in the frontal lobes. The regional and global elevations in white matter T2 suggest abnormalities of white matter tissue water content in autism, which may represent a neurobiological basis for the aberrant cortical connectivity hypothesized to underlie the disorder.

Using the magnetosome to model effective gene‐based contrast for magnetic resonance imaging

Formation of iron biominerals is a naturally occurring phenomenon, particularly among magnetotactic bacteria which produce magnetite (Fe(3) O(4) ) in a subcellular compartment termed the magnetosome. Under the control of numerous genes, the magnetosome serves as a model upon which to (1) develop gene-based contrast in mammalian cells and (2) provide a mechanism for reporter gene expression in magnetic resonance imaging (MRI). There are two main components to the magnetosome: the biomineral and the lipid bilayer that surrounds it. Both are essential for magnetotaxis in a variety of magnetotactic bacteria, but nonessential for cell survival. Through comparative genome analysis, a subset of genes characteristic of the magnetotactic phenotype has been found both within and outside a magnetosome genomic island. The functions of magnetosome-associated proteins reflect the complex nature of this intracellular structure and include vesicle formation, cytoskeletal attachment, iron transport, and crystallization. Examination of magnetosome genes and structure indicates a protein-directed and stepwise assembly of the magnetosome compartment. Attachment of magnetosomes along a cytoskeletal filament aligns the magnetic particles such that the cell may be propelled along an external magnetic field. Interest in this form of magnetotaxis has prompted research in several areas of medicine, including magnetotactic bacterial targeting of tumors, MR-guided movement of magnetosome-bearing cells through vessels and molecular imaging of mammalian cells using MRI, and its hybrid modalities. The potential adaptation of magnetosome genes for noninvasive medical imaging provides new opportunities for development of reporter gene expression for MRI.

Females Follow a More “Compact” Early Human Brain Development Model Than Males. A Case-Control Study of Preterm Neonates

The pattern of sexual differentiation of the human brain is not well understood, particularly at the early stages of development when intense growth and multiple maturational phenomena overlap and interrelate. A case-control study of 20 preterm males and females matched for age was conducted. Three-dimensional images were acquired with 3 T MRI. The cerebral volume and the cortical folding area (FA), defined as the surface area of the interface between cortical gray and white matter, were compared between males and females. Females had smaller cerebra than males even after removing the influence of overall size differences between the subjects. The cortical FA increased in relation to volume by a power of 4/3 in both groups. Females had larger cortical FA compared with males with similar cerebral volumes. The study provides in vivo evidence of sexually dimorphic early human brain development. The relatively more “compact” female model may well relate to sex differences in neural circuitry and cognitive domains.

Biophysical features of MagA expression in mammalian cells: implications for MRI contrast

We compared overexpression of the magnetotactic bacterial gene MagA with the modified mammalian ferritin genes HF + LF, in which both heavy and light subunits lack iron response elements. Whereas both expression systems have been proposed for use in non-invasive, magnetic resonance (MR) reporter gene expression, limited information is available regarding their relative potential for providing gene-based contrast. Measurements of MR relaxation rates in these expression systems are important for optimizing cell detection and specificity, for developing quantification methods, and for refinement of gene-based iron contrast using magnetosome associated genes. We measured the total transverse relaxation rate (R2*), its irreversible and reversible components (R2 and R2′, respectively) and the longitudinal relaxation rate (R1) in MDA-MB-435 tumor cells. Clonal lines overexpressing MagA and HF + LF were cultured in the presence and absence of iron supplementation, and mounted in a spherical phantom for relaxation mapping at 3 Tesla. In addition to MR measures, cellular changes in iron and zinc were evaluated by inductively coupled plasma mass spectrometry, in ATP by luciferase bioluminescence and in transferrin receptor by Western blot. Only transverse relaxation rates were significantly higher in iron-supplemented, MagA- and HF + LF-expressing cells compared to non-supplemented cells and the parental control. R2* provided the greatest absolute difference and R2′ showed the greatest relative difference, consistent with the notion that R2′ may be a more specific indicator of iron-based contrast than R2, as observed in brain tissue. Iron supplementation of MagA- and HF + LF-expressing cells increased the iron/zinc ratio approximately 20-fold, while transferrin receptor expression decreased approximately 10-fold. Level of ATP was similar across all cell types and culture conditions. These results highlight the potential of magnetotactic bacterial gene expression for improving MR contrast.

Changes in Cerebral Oxygen Consumption and High-Energy Phosphates During Early Recovery in Hypoxic-Ischemic Piglets: A Combined Near-Infrared and Magnetic Resonance Spectroscopy Study

Near-infrared spectroscopy (NIRS) offers the ability to assess brain function at the bedside of critically ill neonates. Our group previously demonstrated a persistent reduction in the cerebral metabolic rate of oxygen (CMRO2) after hypoxia-ischemia (HI) in newborn piglets. The purpose of this current study was to determine the causes of this reduction by combining NIRS with magnetic resonance spectroscopy (MRS) to measure high-energy metabolites and diffusion-weighted imaging to measure cellular edema. Nine piglets were exposed to 30 min of HI and nine piglets served as controls. Proton and phosphorous MRS spectra, apparent diffusion coefficient (ADC) maps, and CMRO2 measurements were collected periodically before and for 5.5 h after HI. A significant decrease in CMRO2 (26 ± 7%) was observed after HI. Incomplete recovery of nucleotide triphosphate concentration (8 ± 3% <controls) and reduced ADC (16 ± 5%) suggested mitochondrial dysfunction. However, CMRO2 did not correlate with any metabolite concentration during the last 3 h of the recovery period, and no significant changes were found in phosphocreatine and lactate levels. Therefore, the CMRO2 decrease is likely a combination of impaired mitochondrial function and reduced energy demands during the acute phase, which has been previously observed in the mature brain.

High-contrast 3D neonatal brain imaging with combined T1- and T2-weighted MP-RAGE.

Optimization of magnetization‐prepared rapid gradient‐echo (MP‐RAGE) sequence variations for maximum white matter (WM) versus gray matter (GM) contrast in neonates at 3T was investigated. Numerical simulations were applied to optimize and compare three contrast preparation modules and to assess the effect of phase encoding (PE) order on contrast between WM and thin cortical GM layers. Simulations predict that a new sequence, which combines both T1‐ and T2‐weighting into the contrast preparation and utilizes an interleaved elliptical‐spiral PE order, should provide the strongest contrast between neonatal WM and cortical GM. This sequence was compared to a conventional MP‐RAGE acquisition (i.e., T1‐weighted preparation, centric PE order) for in vivo imaging of seven preterm newborn infants. Regional measurements of the contrast‐to‐noise ratio (CNR) between WM and GM demonstrated an increase of 50–70% (depending on GM region) using the new sequence, in good agreement with theoretical predictions. This improved contrast resulted in superior WM versus GM discrimination in intensity‐based brain tissue segmentations. Magn Reson Med 59:1190–1196, 2008.

A hybrid strategy for correcting geometric distortion in echo-planar images

A hybrid strategy for geometric distortion correction of echo-planar images is demonstrated. This procedure utilizes standard field mapping for signal displacement correction and the so-called reverse gradient acquisition for signal intensity correction. (The term reverse gradient refers to an acquisition of two sets of echo-planar images with phase encoding gradients of opposite polarity.) The hybrid strategy is applied to human brain echo-planar images acquired with and without diffusion-weighting. A comparison of the hybrid distortion corrected images to those corrected with standard field mapping only demonstrates much better performance of the hybrid method. A variant of the hybrid method is also demonstrated which requires the acquisition of only one pair of opposite polarity images within a set of images.

Development of a composite material phantom mimicking the magnetic resonance parameters of the neonatal brain at 3.0 Tesla.

Objective:Development of a composite material phantom, comprised of polyvinyl alcohol cryogel (PVA-C) and an agarose additive, to effectively mimic the magnetic resonance relaxation times (T1 and T2) of neonatal white matter (WM) and gray matter (GM) at 3.0 T. Materials and Methods:Samples of PVA-C with and without agarose were prepared with 1 cycle of freezing/thawing. Measurements of T1 and T2, at 3.0 T, were performed on the samples at temperatures ranging from 20°C to 40°C. Results:A sample temperature of 40°C was required to achieve a T1 value sufficiently long to represent neonatal WM. At this temperature, neonatal WM relaxation times required 3% PVA-C with 0.3% agarose, whereas gray matter relaxation times required 8% PVA-C with 1.4% agarose. Conclusions:By adjusting the sample temperature, polyvinyl alcohol concentration, and agarose concentration, the relaxation times of neonatal brain tissues can be obtained using this composite material.

Optimization of Time-to-peak Analysis for Differentiating Malignant and Benign Breast Lesions with Dynamic Contrast-Enhanced MRI

RATIONALE AND OBJECTIVES The aim of this study was to investigate the feasibility of applying measures sensitive to time-to-peak (T(peak)) heterogeneity as indicators for malignancy on breast dynamic contrast-enhanced magnetic resonance imaging. MATERIALS AND METHODS The study included 39 benign and 97 malignant breast lesions from 103 patients. Lesions were automatically segmented by k-means clustering. Voxel-by-voxel T(peak) values were extracted using an empirical model. The pth percentile values (p = 10, 20…) of the T(peak) distribution within each lesion and the fractional and absolute hot spot volumes were determined, where the hot spot volume is the volume of tissue with T(peak) less than a threshold value. Using the area under the receiver-operating characteristic curve (AUC), these measures were tested as indicators for differentiating fibroadenomas from invasive lesions and from ductal carcinoma in situ, as well as for differentiating nonfibroadenoma benign lesions from these malignant lesions. Region of interest-based T(peak) measurements were also tested. Finally, the relationship between hot spot volume and lesion volume was investigated. RESULTS For differentiating fibroadenomas from malignant lesions, AUC values increased with decreasing values of p. At the optimal threshold (3 minutes), the hot spot volume provided high diagnostic performance (AUC ≥0.96 ± 0.02 for absolute hot spot volume). However, for differentiating nonfibroadenoma benign lesions from malignant lesions, AUC values were low. A significant correlation between absolute hot spot volume and lesion volume was found for malignant lesions and nonfibroadenoma benign lesions. CONCLUSION Quantitative analysis of the T(peak) distribution can be optimized for diagnostic performance, providing indicators sensitive to intralesion T(peak) heterogeneity.

The Interface Between Iron Metabolism and Gene-Based Iron Contrast for MRI

Using a gene-based approach to track cellular and molecular activity with magnetic resonance imaging (MRI) has many advantages. The strong correlation between transverse relaxation rates and total cellular iron content provides a basis for developing sensitive and quantitative detection of MRI reporter gene expression. In addition to biophysical concepts, general features of mammalian iron regulation add valuable context for interpreting molecular MRI predicated on gene-based iron labeling. With particular reference to the potential of magnetotactic bacterial gene expression as a magnetic resonance (MR) contrast agent for mammalian cell tracking, studies in different cell culture models highlight the influence of intrinsic iron regulation on the MRI signal. The interplay between dynamic regulation of mammalian iron metabolism and expression systems designed to sequester iron biominerals for MRI is presented from the perspective of their potential influence on MR image interpretation.