Faculty Bibliography

The mdx mouse has a genetically homologous disease to Duchenne muscular dystrophy in humans. The disease progression, however, is not accompanied by the same level of fatty infiltration and muscle degeneration as occurs in humans. Thus, the presence of histological/pathological changes in living mdx mice has been difficult to monitor. We quantified proton density and the T2 relaxation time of protons with a resolution of 195 x 195 x 1000 microm using multiecho magnetic resonance (MR) imaging at 7 Tesla. These relaxation data were correlated with water content in both muscle and brain of mdx and controls. No differences were observed in brain. The mdx muscles had increased water content and proton density and decreased T2 relative to controls. These data indicate that there are intrinsic changes in T2 (opposite to that which would be induced by fatty infiltration) and suggest that T2 imaging could be used to monitor progression and treatment in this animal model

A simple theoretical model was developed to compare the sensitivities (i.e., signal-to-noise ratios per unit imaging time) of two-dimensional (2D) multislice and 3D imaging sequences. The model shows that the sensitivities of 3D and 2D multislice MRI sequences are usually similar. Sensitivities are identical in T2-weighted sequences when the T(R)s of the two sequences are the same. In T1-weighted gradient-echo sequences, sensitivities are very similar when Ernst angle excitation is used and the T(R) of the 2D sequence is less than T1. The predictions of the model are confirmed in phantom and animal experiments

Blood oxygenation level-dependent (BOLD) magnetic resonance (MR) imaging is sensitive, in part, to the amount of paramagnetic deoxyhemoglobin in a voxel. This project was designed to determine whether there would be differences in the BOLD response between the hippocampus and other brain regions to acute hypoxia. R2* was quantified using a multi-echo gradient-echo sequence. The pyramidal CA1 region of the hippocampus showed a reduced response to changes in arterial oxygenation relative to cortex and basal ganglia and white matter. This difference may relate to the relative hypoxia sensitivity of the hippocampus. It also supports the premise that in functional MR imaging, the magnitude of the MR response to a stimulus may vary with the region of the brain.

RATIONALE AND OBJECTIVES: This study was designed to determine the relative influences of proton density versus collagen fiber orientation (through its influence on T2) in defining the layers of articular cartilage as seen in long-repetition-time magnetic resonance (MR) images. The authors mapped the T2 and proton densities of articular cartilage at 0 degree and 55 degrees with respect to the main magnetic field (B0) to determine the influence of T2 and water content on the normal laminar appearance of hyaline cartilage. MATERIALS AND METHODS: Six patellae of white-tailed deer were imaged at 7 T. T2 and proton densities were calculated from echo time versus signal intensity plots obtained with a multiecho, composite pulse sequence. Regions of interest in the radial and transitional zones were compared with the articular facets at 0 degree and 55 degrees relative to B0. Transmission electron microscopy was performed for correlation. RESULTS: At 0 degree, T2 was longer in the transitional than in the radial zone (29 vs 11 msec). AT 55 degrees, T2 increased in both radial and transitional zones, although the difference between the zones decreased (37 vs 29 msec). There was no difference in proton density between the two layers. CONCLUSION: Collagen fiber orientation, through T2 effects, is the dominant influence on the appearance of layers in hyaline cartilage in long-repetition-time MR images; proton density is not a major factor, and the collagen fiber orientation in the transitional zone is not totally random

An optimization scheme was developed for gradient echo imaging using a half-birdcage RF coil at 7 T to obtain maximal contrast between gray and white matter in the spinal cord of rodents. This optimization was combined with microimaging techniques to obtain in vivo pixel sizes of 78 x 78 x 700 microm. These techniques can be implemented in an in vivo study to investigate the myelin structure within the white matter of the rodent spinal cord

This paper will discuss the assumptions, strengths and weaknesses of both BOLD imaging and NIR spectrophotometry with respect to monitoring tissue oxygenation. BOLD, or blood oxygen level dependent MRI, is an imaging protocol that is sensitive to specific relaxation rates which are influenced by deoxyhemoglobin. NIRS is capable of providing information on oxyhemoglobin, deoxyhemoglobin and total hemoglobin. Both techniques have inherent assumptions, strengths and weaknesses. NIRS has not been able to provide the spatial sensitivity of BOLD. BOLD may be more difficult to quantify. Currently, these two methods are complementary, providing data that strengthens the interpretation of results from each modality. Recent data indicate that relaxation rates can be used to quantify deoxyhemoglobin in isolated blood and, under specific conditions, there is a strong correlation between deoxyhemoglobin content measured by NIRS and changes in relaxation rates measured by MRI. These data indicate that BOLD imaging has the potential to become an attractive alternative to NIRS

Improvements of the Cadzow enhancement procedure (EP) are reported. A numerical study showed that once EP has properly converged, it results in a near-optimum signal fitting the noisy data in the least-squares sense. At convergence, the estimated signal involves purely exponentially damped sinusoids; then any linear-prediction methods can be used indifferently in the subsequent quantitation step. Moreover, this final step needs only a few samples. Amplitude estimates have negligible bias even for low SNR and their standard deviations are nearly equal to their Cramer-Rao lower bounds. This is illustrated in a Monte Carlo simulation and an in vitro study using three different SVD-based quantitation methods (LPSVD, HSVD, TLS). EP used prior to any linear-prediction methods leads to an important reduction of the threshold SNR and results in near-optimal automatic quantitation methods. (C) 1994 Academic Press, Inc.