확산 텐서 영상을 이용한 정상 뇌백질과 뇌종양의 분할 비등방도 측정

Abstract

Purpose: The purpose of this study was to measure the apparent diffusion coefficient (ADC) and fractional anisotropy (FA) of normal adult brain tissue and tumors, and to compare the differences. Materials and Methods: Eight normal adults and ten patients in whom intracranial tumors had been diagnosed were included. Imaging was performed using a 1.5 T MR unit and a single-shot spin-echo EPI pulse sequence (TR/TE=4024/94 msec, 128 acquisition/256 reconstruction, 23 cm FOV, 5 mm thickness, 2 mm interslice gap, 4 NSA), six different direction gradients (x, y, z, xy, yz, xz), and 2 b-values (0, 1000). Isotropic ADC (D) was obtained from seven images per slice, and fractional anisotropy (FA) was calculated from the isotropic ADC and eigenvalues of three directions. A region of interest was drawn at frontal gray and white matter, periventricular white matter, the corpus callosum, internal capsule, caudate nucleus and center of the tumor mass, and for each region, fractional anisotropy readings were obtained. Results: In normal adults, the findings were as follows: frontal gray matter: D=0.81±0.06, FA=0.32±0.03; frontal white matter: D=0.79±0.04, FA=0.56±0.09, periventricular white matter: D=0.77±0.02, FA=0.51 ±0.04; corpus callosum: D=0.79±0.07, FA=0.82±0.07; internal capsule: D=0.73±0.04, FA=0.77±0.05; caudate nucleus: D=0.76±0.05, FA=0.35±0.05. High anisotropy was demonstrated in white matter, especially in the corpus callosum and internal capsule, and the degree of anisotropy was similar in gray and deep gray matter. For most brain tumors, isotropic ADC was similar to that of white matter, but fractional anisotropy was lower. A low-grade astrocytoma showed higher isotropic ADC and lower fractional anisotropy than normal white matter, and at the center of a meningioma, fractional anisotropy was high. Conclusion: For the classification of brain tumors and determination of the extent of disease, comparison between the apparent diffusion coefficient and fractional anisotropy is useful.