General Brain Imaging: Technique Development

Currently many structural neuroimaging studies rely only on a T1-weighted image for brain extraction. Additional image contrast like T2 may improve the performance of the automatic brain extraction procedure. In this work, a previously proposed multiparametric 3D structural imaging sequence that provides several volumes with varying contrast in a multi-echo acquisition is used to assist in automatic brain segmentation. Two 3D volumes (one T1-weighted and one T2-weighted) with 1.2 mm isotropic resolution and a low resolution 3D field map were obtained simultaneously within 6.5 minutes. Improvement in brain extraction utilizing the additional contrast was observed.

A novel parallel imaging technique for PROPELLER that utilizes external calibration data as well as a small amount of internal calibration data per blade is demonstrated for T1 FLAIR imaging. Short echo trains, which are optimal for T1 imaging, are maintained while the effective blade width is increased by reducing the number of internal calibration lines. Wider blades enable motion to be detected and corrected more reliably, improving robustness in uncooperative patients. The method is validated in studies of volunteers instructed to move their head during the acquisition.

The MR flow quantification of the vessels can be used to evaluate the response of the vessel-related surgeries. However, for the conventional longitudinal studies of blood flow, manual slice selection in different days can cause measurement variation and thus degrade the accuracy of evaluation. In our study, in order to solve this problem, an automatic slice positioning method combined with previously implemented automatic ROI selection method were proposed to reduce the inter-scan variation. Furthermore, GPU-accelerated computation was applied to speed up the image registration. The flow variation of the human study was reported.

The human brain is extremely sensitive and vulnerable to even small alterations in oxygen supply, making a measure for assessing global cerebral metabolic rate of oxygen consumption (CMRO2) very important. We propose a method for estimating CMRO2 by simultaneous quantification of oxygen saturation by MR oximetry and cerebral blood flow by phase-contrast MRI in the major vessels draining (superior sagittal sinus) and feeding the brain (internal carotid and vertebral arteries), respectively. Our results demonstrate that the proposed technique is robust and reproducible, yielding temporally stable measurements at a temporal resolution 30 seconds.

Inclusion of fat suppress (FS) pulses in a spin echo acquisition substantially reduces scan efficiency. At 3 Tesla, the incidental magnetization transfer and dielectric effects from the FS pulses also degrade the image contrast and image uniformity. In this study, we developed a spin echo two-point Dixon technique for fat-suppressed T1-weighted imaging. In comparison to the conventional spin echo with FS pulses, we demonstrated in vivo that the new technique was 40% more efficient and had much better image contrast, better FS and overall image uniformity for fat-suppressed T1-weighted imaging of head & neck at 3 Tesla.

We aimed to design a head holder for calculation the susceptibility through multiple orientation sampling (COSMOS). Different brain tissues with different susceptibility result in the changes of focal magnetic fileds. Quantitative susceptibility imaging of brain are obtained by measurement of the focal magnetic field changes.. It is a promising approach for exploring various brain pathological conditions. For clinical application, the challenging problem is to rotate the head of patient along the Y axis only and fix the head in the degree through the MR scan. Our results show great control of the rotation in three axes with the facilitation of head holder.

The MNI standard template is ideal for healthy subjects. To improve the accuracy of normalization and the further analysis (such as statistical analysis), we created a study specific template in MNI space for an aged population with mild to critical aortic stenosis. This study specific template is approximately in MNI template space, but accommodates some characteristics of this particular study population, especially incorporating the increased ventricular volume.

The use of a head gradient set within a wholebody magnet can lead to significant artifacts. Signal arising from the shoulders is encoded in such a way that it aliases into the main image. Such artifacts are particularly pronounced at higher field, where B1 effects cause the sensitive region of volume coils to extend out into the chest and shoulder region. Previous approaches have tried to diminish the RF penetration in this area (through the use of RF shielding materials) or to disrupt the local field through using ferromagnetic material shown into a jacket that the subject wears. This study shows that with the use of both the wholebody and head gradient shim sets the signal from the shoulders could be dephased without affecting the signal from the head itself.

Phased-array coils with increasing number of array elements have been developed and employed for clinical MR imaging. A phased-array head coil with 32 array elements provides quantitatively better brain images as compared to commercially available head coil with lower number of array elements. In this study, we investigate if application of this coil would improve the anatomic and pathologic analysis from the resultant brain images by qualitatively comparing MR images obtained using this coil with those obtained using a commercially available phased-array head coil with 12 array elements.

The objective of this study was to measure the cross relaxation exchange time between the myelin water and intra/extracellular water pools in healthy human white matter in vivo. Five different white matter structures were investigated. Bloch equations were solved analytically and cross relaxation exchange times were extracted. Due to the ambiguity in the literature on spin-lattice relaxation times in white matter, three T1 scenarios were developed. The extracted cross relaxation times were then used to estimate the exchange corrections for myelin water fraction (MWF) measurements. The choice of T1 scenario had a significant effect on cross relaxation times and consequently on MWF corrections.

An agarose gel and saline solution phantom was developed to mimic properties of the human brain at 7T. This design provides many parameters matched to the behaviour seen in vivo including: grey matter/white matter contrast for sequence development, matched B1+ interference (RF wavelength) behaviour, and coil loading effects.

The light microscopy images have a high spatial resolution and are usually co-registered to the corresponding MR images in order to make the comparison. However, there are artifacts, such as tearing, deformation, or disappearance of tissue fragments, in the stained slices. Those artifacts make the registration among MR, blockface, and light images more difficult. In this study, two image post-processing techniques are introduced, which can provide a better initialization to the nonrigid registration algorithm.

We determined the inter-reader reproducibility in the assessment of carotid territory brain lesions. In a carotid study, two reviewers reviewed brain images of 134 hemispheres from 67 subjects independently. The inter-reader agreement was substantial for the lesion presence (κ = 0.67; 0.54-0.80) and the lesion size (κ = 0.75; 0.60-0.92), and there was complete agreement (κ = 1.0) for age. Some disagreements on the presence or absence of lesions may be due to their location near the boundary of the carotid territory. It is important to point out small old lesions were the primary factor that reduced reproducibility.

Cerebral oxygen extraction factor (OEF) provides critically important information to assess the brain oxygen metabolism in both normal and disease states. It has been reported that the OEF could be evaluated for healthy human volunteers using MRI. In this study, a based gradient echo sampling of the spin echo (GESSE) sequence implemented on GE 3.0T scanner to evaluate the cerebral OEF distribution of rabbits before and after carotid artery occlusion operation. Our study demonstrated a consistent and significant increase of OEF in rabbits post carotid artery occlusion, suggesting that this MR based method can be utilized to detect pathophysiological changes in cerebral oxygenation.

Magnetic resonance imaging provides a mean to non-invasively investigate the neurological cause of dyslexia, a learning disability that affects one’s ability to read despite adequate intelligence. Myelin water imaging, based on calculating the myelin water fraction from multi-echo T2 relaxation curves, can be used to quantitatively assess white matter. In this study, myelin water imaging and cognitive and reading assessments were performed on children with a wide range of reading ability in order to investigate the relationship between dyslexia and white matter development in the brain.

There is no treatment of TBI partially due to the current clinical classification system cannot effectively identify the pathoanatomical information of the brain. We propose that a synergistic use of multi-imaging techniques may capture much of the heterogeneity and complexity of brain injury in individual patients therefore result in improved accuracy and detail in prognostic models and improved efficiency of clinical trials. In this study, we report our preliminary observations regarding the synergistic use of these three MRI techniques in an improved characterization of TBI.

Time resolved 3D-phase contrast imaging allows to assess cerebrospinal fluid pulsation at the craniocervical junction and cervical spine. Using state-of-the-art visualization techniques it helps to identify and differentiate between pathological and physiological cerebrospinal fluid flow pattern. As such it may add valuable information for the analysis of pathologies associated with altered cerebrospinal fluid flow like in Chiari malformations and may promote a better understanding of the underlying pathophysiology of these diseases.

Subdural haematoma (SDH) is an emergency when acute, and management is often guided by imaging. Distinguishing between the different phases of SDH may be possible by the presence and position of methaemoglobin. In this study, we create an in vitro system of SDH and examine the relationship between intra- and extracellular methaemoglobin at various concentrations, using R2, R2* and R2’ parameters. R2’ relaxation rate appears to be the most sensitive marker of methaemoglobin concentration and can readily differentiate between when it is intra- or extracellular. This parameter could therefore be used to stage the phase of SDH.

The carotid MR hyperintense (vulnerable) plaque is associated with a higher morbidity and mortality. We investigated the prevalence of this plaque and its clinical correlates on T₁-weighted turbo-field echo (T₁w-TFE) MRI in patients with ischemic symptoms and varying degrees of stenosis. 153 patients with TIA or ischemic infarct were retrospectively examined. 18% showed one or more hyperintense plaques. Half of all hyperintense plaques occurred in symptomatic patients with either 0-69% stenosis or occlusion; more than ⅓ of patients with 50-69% stenosis presented with this plaque. This subgroup of patients could in future possibly benefit from aggressive medicinal therapy or revascularization.

Susceptibility-Weighted Imaging (SWI) improves the visibility of target structures (globus pallidus (GP) and subthalamic nucleus (STN)) for deep brain stimulation (DBS). However, because phase contrast is non-local and orientation dependent, SWI contains artifacts that may result in targeting errors. Susceptibility maps, which have shown promise for overcoming such artifacts, were calculated from clinical 1.5-T phase data acquired at a single orientation. 2-mm shifts in the superior borders of the red nuclei and GP in the SWI relative to the susceptibility maps were observed in several volunteers and DBS patients showing that susceptibility mapping may help reduce SWI targeting errors.

Aqueductal cerebrospinal fluid flow, measured with motion sensitive phase contrast MRI, is subject to typical imperfections in flow imaging such as partial volume effects and difficulties in lumen delineation. In this study we abandoned graphically represented vessel delineations. Instead we used complex data generated by the PC-MRI in a method with the potential of objective and absolute measurement of CSF velocity and area, without restrictions by matrix resolution. We compared the complex value methodology with conventional manual segmentation for the determination of aqueduct CSF stroke volume in a group of 42 healthy elderly.

In this study high field MR and phase contrast (PCMR) data were used to construct a subject-specific model of flow in the vertebro-basilar system. Wall shear stress (WSS) data from the model yield insight into artery bypass graft design.