Zhengrong Liang PhD, PRESENTER: Nothing to Disclose

Abstract: HTML Purpose: Cognitive impairment (CI) is commonly presented in neural disorders. 1H MRS has provided quantitative information about major pathologic aspects. However, it does not enable us to obtain information on specific brain tissues. On the other hand, volumetric MRI analyses have provided morphological processes of brain tissues. The goal of this study is to map MR spectroscopy of interested chemical compounds onto their corresponding cerebral tissues and establish an effective means to explore the complementary information between these two measures. Methods and Materials: Multi-voxel 1H MRS data were acquired with a 1.5T Marconi scanner with a PRESS sequence, TE = 135 ms, TR = 1500 ms, 16 cm FOV, 2D phase encoding (16x16), and 2 scan averages. The slice of interest with 2 cm thickness was taken through the posterior and anterior aspects of the corpus callosum. For each MRS voxel (1x1x2 cm3), three peaks of N-acetyl aspartate (NAA), choline compounds (Cho), and total creatine (Cr) were digitized. That is, each MRS voxel had three values, from which a ratio of NAA over Cho or NAA over Cr was computed for a relative measure of NAA level. Multispectral (T1, T2, and FLAIR) MR images were acquired with the same scanner with 256x256 array and voxel size of 0.9x0.9x1.5 mm3. A maximum a posteriori probability expectation-maximization framework has been established to segment the cerebral tissue components in each voxel. With the segmented tissue fractions/mixtures in each image voxel, the metabolite changes of the chemical compounds inside the cerebral tissues, measured by 1H MRS, can be mapped onto the image voxels for any correlation study. Assume that WM has a weight factor of w contributing to the NAA measure, GM with a factor of g, CSF with a factor of c = 0 and lesion with a factor of s. Then, for each MRS voxel, we have NAA = wWM + gGM + sLesion. Considering neighborhood correlation among MRS voxels, the mapping becomes a back-projection operation in tomographic imaging. Because both MRI and MRS images are spatially registered, a correlation between atrophy and NAA can be performed globally and locally across the FOV. Results: Six MS patients (20-55 years old) with cognitive impairment were recruited. MRS images were mapped onto the segmented tissue fractions for localized measures of metabolic changes. The correlation of the mapped images on the measures of NAA and cerebral tissue volume is r=0.53 (p=0.000). Conclusion: The integrated MRI/MRS framework opens a new way to explore the MRS metabolic changes and MRI morphological processes for studying brain functions.       Questions about this event email: lli@mil.sunysb.edu

Liang PhD, Z, An Integrated MRI and MRS Framework for Studies of Cognitive Impairment.  Radiological Society of North America 2003 Scientific Assembly and Annual Meeting, November 30 - December 5, 2003 ,Chicago IL. http://archive.rsna.org/2003/3108332.html