[국내논문]화소 분석의 최적화를 위해 자화감수성 영상에 나타난 뇌조직의 가우시안 필터 효과 연구 Gaussian Filtering Effects on Brain Tissue-masked Susceptibility Weighted Images to Optimize Voxel-based Analysis원문보기
목적: 본 연구의 목적은 자화감수성 영상 (SWI)에 나타난 정상 노인의 뇌조직을 픽셀 별로 분석하기 위해 사용되는 다듬질 (smoothing)의 핵심 크기 효과를 보는 것이다. 대상과 방법: 이십 명의 정상 지원군 (평균 나이${\pm}$표준 편차 = $67.8{\pm}6.09$세, 여 14명, 남 6명) 이 실험에 대한 동의와 함께 본 연구에 참여하였다. 이 지원군 각각의 자화감수성 영상을 만들기 위해 일차원 혈류흐름 보상 삼차원 경사자장 에코 시퀀스를 이용해 크기과 위상 영상을 얻었고, 영상 처리와 영상 내 조직 분할에 사용되는 자화준비 급속획득 경사자장 에코 (MPRAGE) 시퀀스를 이용한 삼차원 시상면 T1 강조영상을 얻었다. 자화감수성 영상은 다시 위상영상을 이용하여 상자성 (paramagnetic) 물질의 존재 여부를 강조하는 PSWI (위상 영상에서 양수 값을 강조한 자화감수성 영상)과 반자성 (diamagnetic) 물질의 존재 여부를 강조하는 NSWI (위상 영상의 음수 값을 강조한 자화감수성 영상) 영상을 만들었다. 오직 뇌조직 부분만 나타나도록 조직이 아닌 부분을 차폐 (masking) 하는 과정을 거쳤다. 마지막으로 뇌조직 PSWI와 NSWI는 등방성의 0, 2, 4, 8 mm의 다듬질 핵심 크기를 이용하여 다듬질 되었다. 또한 각각의 다듬질 핵심 크기로 다듬질된 PSWI와 NSWI를 쌍 비교 t검정을 실행하여 각 픽셀 별로 비교하였다. 결과: 통계 분석의 중요도는 다듬질의 핵심 크기가 커질수록 증가하였고, 영상의 시그널 세기는 NSWI가 PSWI보다 컸다. 또한 영상의 픽셀 별 비교 분석에 가장 최적화 된 다듬질의 핵심 크기는 4였으며 쌍 비교 t검정 결과 뇌의 양쪽에서 차이가 난 뇌 조직의 위치와 범위는 뇌의 여러 지역에서 발견되었다. 결론: 상자성 물질을 강조한 PSWI는 자화감수성이 높은 뇌 여러 영역의 시그널 크기를 감소시켰다. 부분적인 부피효과와 큰 혈관의 기여도를 최소화 하기 위해서는 뇌 조직만 뽑아낸 자화감수성 영상의 복셀 별 분석이 사용되어야 하겠다.
목적: 본 연구의 목적은 자화감수성 영상 (SWI)에 나타난 정상 노인의 뇌조직을 픽셀 별로 분석하기 위해 사용되는 다듬질 (smoothing)의 핵심 크기 효과를 보는 것이다. 대상과 방법: 이십 명의 정상 지원군 (평균 나이${\pm}$ 표준 편차 = $67.8{\pm}6.09$세, 여 14명, 남 6명) 이 실험에 대한 동의와 함께 본 연구에 참여하였다. 이 지원군 각각의 자화감수성 영상을 만들기 위해 일차원 혈류흐름 보상 삼차원 경사자장 에코 시퀀스를 이용해 크기과 위상 영상을 얻었고, 영상 처리와 영상 내 조직 분할에 사용되는 자화준비 급속획득 경사자장 에코 (MPRAGE) 시퀀스를 이용한 삼차원 시상면 T1 강조영상을 얻었다. 자화감수성 영상은 다시 위상영상을 이용하여 상자성 (paramagnetic) 물질의 존재 여부를 강조하는 PSWI (위상 영상에서 양수 값을 강조한 자화감수성 영상)과 반자성 (diamagnetic) 물질의 존재 여부를 강조하는 NSWI (위상 영상의 음수 값을 강조한 자화감수성 영상) 영상을 만들었다. 오직 뇌조직 부분만 나타나도록 조직이 아닌 부분을 차폐 (masking) 하는 과정을 거쳤다. 마지막으로 뇌조직 PSWI와 NSWI는 등방성의 0, 2, 4, 8 mm의 다듬질 핵심 크기를 이용하여 다듬질 되었다. 또한 각각의 다듬질 핵심 크기로 다듬질된 PSWI와 NSWI를 쌍 비교 t검정을 실행하여 각 픽셀 별로 비교하였다. 결과: 통계 분석의 중요도는 다듬질의 핵심 크기가 커질수록 증가하였고, 영상의 시그널 세기는 NSWI가 PSWI보다 컸다. 또한 영상의 픽셀 별 비교 분석에 가장 최적화 된 다듬질의 핵심 크기는 4였으며 쌍 비교 t검정 결과 뇌의 양쪽에서 차이가 난 뇌 조직의 위치와 범위는 뇌의 여러 지역에서 발견되었다. 결론: 상자성 물질을 강조한 PSWI는 자화감수성이 높은 뇌 여러 영역의 시그널 크기를 감소시켰다. 부분적인 부피효과와 큰 혈관의 기여도를 최소화 하기 위해서는 뇌 조직만 뽑아낸 자화감수성 영상의 복셀 별 분석이 사용되어야 하겠다.
Purpose : The objective of this study was to investigate effects of different smoothing kernel sizes on brain tissue-masked susceptibility-weighted images (SWI) obtained from normal elderly subjects using voxel-based analyses. Materials and Methods: Twenty healthy human volunteers (mean $age{\p...
Purpose : The objective of this study was to investigate effects of different smoothing kernel sizes on brain tissue-masked susceptibility-weighted images (SWI) obtained from normal elderly subjects using voxel-based analyses. Materials and Methods: Twenty healthy human volunteers (mean $age{\pm}SD$ = $67.8{\pm}6.09$ years, 14 females and 6 males) were studied after informed consent. A fully first-order flow-compensated three-dimensional (3D) gradient-echo sequence ran to obtain axial magnitude and phase images to generate SWI data. In addition, sagittal 3D T1-weighted images were acquired with the magnetization-prepared rapid acquisition of gradient-echo sequence for brain tissue segmentation and imaging registration. Both paramagnetically (PSWI) and diamagnetically (NSWI) phase-masked SWI data were obtained with masking out non-brain tissues. Finally, both tissue-masked PSWI and NSWI data were smoothed using different smoothing kernel sizes that were isotropic 0, 2, 4, and 8 mm Gaussian kernels. The voxel-based comparisons were performed using a paired t-test between PSWI and NSWI for each smoothing kernel size. Results: The significance of comparisons increased with increasing smoothing kernel sizes. Signals from NSWI were greater than those from PSWI. The smoothing kernel size of four was optimal to use voxel-based comparisons. The bilaterally different areas were found on multiple brain regions. Conclusion: The paramagnetic (positive) phase mask led to reduce signals from high susceptibility areas. To minimize partial volume effects and contributions of large vessels, the voxel-based analysis on SWI with masked non-brain components should be utilized.
Purpose : The objective of this study was to investigate effects of different smoothing kernel sizes on brain tissue-masked susceptibility-weighted images (SWI) obtained from normal elderly subjects using voxel-based analyses. Materials and Methods: Twenty healthy human volunteers (mean $age{\pm}SD$ = $67.8{\pm}6.09$ years, 14 females and 6 males) were studied after informed consent. A fully first-order flow-compensated three-dimensional (3D) gradient-echo sequence ran to obtain axial magnitude and phase images to generate SWI data. In addition, sagittal 3D T1-weighted images were acquired with the magnetization-prepared rapid acquisition of gradient-echo sequence for brain tissue segmentation and imaging registration. Both paramagnetically (PSWI) and diamagnetically (NSWI) phase-masked SWI data were obtained with masking out non-brain tissues. Finally, both tissue-masked PSWI and NSWI data were smoothed using different smoothing kernel sizes that were isotropic 0, 2, 4, and 8 mm Gaussian kernels. The voxel-based comparisons were performed using a paired t-test between PSWI and NSWI for each smoothing kernel size. Results: The significance of comparisons increased with increasing smoothing kernel sizes. Signals from NSWI were greater than those from PSWI. The smoothing kernel size of four was optimal to use voxel-based comparisons. The bilaterally different areas were found on multiple brain regions. Conclusion: The paramagnetic (positive) phase mask led to reduce signals from high susceptibility areas. To minimize partial volume effects and contributions of large vessels, the voxel-based analysis on SWI with masked non-brain components should be utilized.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
The smoothing kernel size should be optimized to increase the validity of parametric statistical tests. The objective of this study was to investigate effects of different smoothing kernel sizes on brain tissue-masked SWI to enhance the brain tissue contrast rather than the vessel contrast using voxel-based analyses. The smoothing kernel sizes were investigated by varying the smoothing factors during voxel-wise comparisons with masking brain tissues on SWI using segmented brain tissues obtained from 3D T1-weighted images.
One limitation of this study was the uncertain relationship between phase signals and the susceptibility effects. Although phase images contain information on susceptibility differences, the phase contrast does not fully correspond with the susceptibility differences; therefore, one cannot easily conclude that the final SWIs produced by the phase mask multiplication purely enhanced the susceptibility effects.
The main objective of this study was to investigate voxel-based differences of signal intensities of the paramagnetic phase-masked SWI data and the diamagnetic phase-masked SWI data with varying the smoothing kernel sizes. The smoothing kernel size of four could be optimal to use voxel-based comparisons on SWI data.
제안 방법
09, range = 62 to 80, 14 females and 6 males) with no medical history of neurological diseases were studied after informed consent under the institutional review board-approved protocol. MR imaging was performed on a 3T clinical MR system (Achieva, Philips Medical Systems, Best, The Netherlands) equipped with an eight-channel sensitivity encoding head coil. A fully first-order flow-compensated 3D echo-shifted gradient-echo sequence ran to obtain axial magnitude and phase images to generate SWI data.
Because the diamagnetic phase mask enhanced a few soft tissues while suppressing vein signals, it is assumed to reduce the susceptibility effects revealed in the SWI. The magnitude images were then multiplied by the paramagnetic or diamagnetic phase masks four times to create positively phase-masked SWI (PSWI) and negatively phase-masked SWI (NSWI), respectively. The phase multiplication of four produced the most optimal contrast-to-noise ratio in SWI (3).
). For each subject, the 3D T1-weighted images were overlaid with the magnitude images of the 3D gradient-echo images for coregistration and were then spatially normalized to a standard 3D T1-weighted brain template, which was created by averaging 123 brains of the elderly (mean age = 68.2, SD = 8.6) using a 12-parameter nonlinear transformation (8). The transformation parameter obtained from the 3D T1-weighted image for each subject was applied to normalize magnitude, PSWI, and NSWI which were interpolated to the 1 × 1 × 1 mm3 voxel size.
0. After the masking process, both brain-tissue masked PSWI and NSWI data were smoothed using four different smoothing kernel sizes that were isotropic 0 mm, 2 mm, 4 mm, or 8 mm Gaussian kernels to investigate effects of smoothing on voxel-based comparisons.
All voxel-based statistical analyses were achieved using the SPM5 software. In order to investigate paramagnetic and diamagnetic phase masks with different smoothing kernel sizes, the voxel-based comparisons of the smoothed PSWI and NSWI data were performed using the paired t-test between PSWI and NSWI for each smoothing kernel size. The gender and age information of each subject was included as covariates.
When the kernel size was too large, on the other hand, the results may show regions that are not so statistically significant. The Gaussian filter of 2 times or 3 times greater than a voxel size is applicable for analyzing 3DT1 weighted images, but the kernel size of eight, which is eight times greater than a voxel size is too great for perform a voxel-based analysis of SWI. Therefore, the voxel-based analysis on SWI data should be carefully used with the optimized steps that should be included in the generation of the SWI brain template, normalization, and in the spatial smoothing.
On the paramagnetic phase-masked SWI data, signal losses were usually found on the venous vessels. In this study, this contribution was minimized by applying brain tissue masks. The differences between PSWI and NSWI should be caused by susceptibility differences in the brain tissues rather than the venous blood vessels.
In this study, we showed possibility of applications of voxel-based analysis of SWI data to investigate human brains in cortex and subcortex areas. In addition, the results of ROI-based analysis supported the results of the voxel-based analysis.
Therefore, the future studies on magnetic susceptibility using SWI should be benefited from these quantitative susceptibility maps, which would exclusively reveal the susceptibility values within the voxels. Moreover, the long acquisition time for a 3DT1 weighted image in addition to SWI is required to perform the image analysis. Finally, the central filter size of 64 × 64 and the phase multiplication of four were applied in this study as suggested in the previous study by Haacke et al.
이론/모형
To obtain brain-tissue masked SWI, the coregistration and normalization steps were achieved using a statistical parametric mapping-version 5 (SPM5) program (Wellcome Department of Imaging Neuroscience, University College, London, U.K.). For each subject, the 3D T1-weighted images were overlaid with the magnitude images of the 3D gradient-echo images for coregistration and were then spatially normalized to a standard 3D T1-weighted brain template, which was created by averaging 123 brains of the elderly (mean age = 68.
참고문헌 (16)
Rauscher A, Sedlacik J, Barth M, Mentzel HJ, Reichenbach JR. Magnetic susceptibility-weighted MR phase imaging of the human brain. AJNR Am J Neuroradiol 2005;26:736-742
Haacke EM, Cheng NY, House MJ, et al. Imaging iron stores in the brain using magnetic resonance imaging. Magn Reson Imaging 2005;23:1-25
Haacke EM, Mittal S, Wu Z, Neelavalli J, Cheng YC. Susceptibility-weighted imaging: technical aspects and clinical applications, part 1. AJNR Am J Neuroradiol 2009;30:19-30
Hagberg GE, Welch EB, Greiser A. The sign convention for phase values on different vendor systems: definition and implications for susceptibility-weighted imaging. Magn Reson Imaging 2010;28:297-300
Eissa A, Lebel RM, Korzan JR, et al. Detecting lesions in multiple sclerosis at 4.7 tesla using phase susceptibility-weighting and t2-weighting. J Magn Reson Imaging 2009;30:737-742
Grabner G, Dal-Bianco A, Schernthaner M, Vass K, Lassmann H, Trattnig S. Analysis of multiple sclerosis lesions using afusion of 3.0 t flair and 7.0 t swi phase: flair swi. J Magn Reson Imaging 2011;33:543-549
Kim MJ, Jahng GH, Lee HY, et al. Development of a Korean standard structural brain template in cognitive normals and patients with mild cognitive impairment and Alzheimer's disease. J. Korean Soc. Magn. Reson. Med. 2010;14:103-114
Fukunaga M, Li TQ, van Gelderen P, et al. Layer-specific variation of iron content in cerebral cortex as a source of MRI contrast. Proc Natl Acad Sci U S A 2010;107:3834-3839
Duyn JH, van Gelderen P, Li TQ, de Zwart JA, Koretsky AP, Fukunaga M. High-field MRI of brain cortical substructure based on signal phase. Proc Natl Acad Sci U S A 2007;104:11796- 11801
Thomas B, Somasundaram S, Thamburaj K, et al. Clinical applications of susceptibility weighted MR imaging of the brain - a pictorial review. Neuroradiology 2008;50:105-116
Niwa T, Aida N, Kawaguchi H, et al. Anatomic dependency of phase shifts in the cerebral venous system of neonates at susceptibility- weighted MRI. J Magn Reson Imaging 2011;34:1031- 1036
Schafer A, Wharton S, Gowland P, Bowtell R. Using magnetic field simulation to study susceptibility-related phase contrast in gradient echo MRI. Neuroimage 2009;48:126-137
Shmueli K, de Zwart JA, van Gelderen P, Li TQ, Dodd SJ, Duyn JH. Magnetic susceptibility mapping of brain tissue in vivo using MRI phase data. Magn Reson Med 2009;62:1510- 1522
de Rochefort L, Brown R, Prince MR, Wang Y. Quantitative MR susceptibility mapping using piece-wise constant regularized inversion of the magnetic field. Magn Reson Med 2008;60:1003- 1009
Liu T, Spincemaille P, de Rochefort L, Kressler B, Wang Y. Calculation of susceptibility through multiple orientation sampling (cosmos): a method for conditioning the inverse problem from measured magnetic field map to susceptibility source image in MRI. Magn Reson Med 2009;61:196-204
※ AI-Helper는 부적절한 답변을 할 수 있습니다.