뇌자기공명영상에서 GRASE 기법과 Turbo Spin Echo 기법의 비교 = Comparison of GRASE and Turbo Spin Echo Sequences in Brain MR Imaging

다국어 초록 (Multilingual Abstract)

Purpose : The purpose of this study was to evaluate the utility of GRASE images of the brain, which combine gradient echo-EPI and turbo spin echo, and to compare the results with those of the turbo spin echo (TSE) technique. Materials and Methods : We analyzed and compared brain MR images obtained in 25 patients between October, 1996 and January, 1997, both TSE and GRASE techniques. Diagnosis was normal (n=5), infarct or ischemia(n=10), intracerebral hemorrhage(n=6), and neoplasm(n=4). TSE and GRASE MR images were obtained using a1.5T Gyroscan ACS-NT(Philips, Netherlands). For qualitative assessment, overall image quality, discrimination between cortical gray-white matter and basal ganglia-white matter, lesion conspicuity, and MR artifact were evaluated using a subjective grading system ranging from 1 to 5(1=TSE better than GRASE, 5=GRASE better than TSE). For quantitative assessment, signal-to-nosise ratios (SNRs) were calculated for cortical gray matter, white matter, basal ganglia, and lesions, and contrast-to-nosise ratios (CNRs) were calculated for cortical gray-white matter, basal ganglia-white matter and lesions-white matter, We measured image acquisition time and RF specific absorption rate (SAR) on TSE and GRASE. Results : Qualitative assessment showed that overall image quality, discrimination between cortical gray-white matter and basal ganglia-white matter, and lesion conspicuity were not significantly different between the two imaging techniques. MR artifact was more frequently seen on GRASE than on TSE. Quantitative assessment showed that the SNR of each brain tissue and lesion was significantly greater on TSE images than on GRASE. Cortical gray-white matter and basal ganglia-white matter CNRs were significantly greater on GRASE images than on TSE, but lesion-white matter CNR was not significantly different between the two techniques. Image acquisition time was 3 minute 31 second in TSE and 1 minute 12 second in GRASE. SAR was lower in GRASE than in TSE. Conclusion : With regard to MR artifact, GRASE is more sensitive than TSE, but as regards image quality and lesion distinction, the two modalities show no distinct difference. Since GRASE is superior to TSE for the detection of hemosiderin, and image acquisition time is three times shorter using GRASE, GRASE might usefully be applied during the evaluation by MR imaging of certain brain conditions.