Schwab Nadine A, Tanner Jared J, Nguyen Peter T, Schmalfuss Ilona M, Bowers Dawn, Okun Michael, Price Catherine C
Department of Clinical and Health Psychology, University of Florida, 1225 Center Drive, Gainesville, FL, 32611, USA.
Department of Radiology, North Florida/South Georgia Veteran Administration, University of Florida, 1601 SW Archer Rd., Gainesville, FL, 32608, USA.
Brain Imaging Behav. 2015 Dec;9(4):744-53. doi: 10.1007/s11682-014-9332-x.
Brain magnetic resonance image (MRI) registration alters structure orientation, size, and/or shape. To determine whether linear registration methods (image transformation to 6, 9, and 12° of freedom) alter structural volume and cognitive associations, we examined transformation alterations to the caudate nucleus within individuals diagnosed with Parkinson's disease (PD) and demographically matched non-PD peers. Volumes from native and six were expected be significantly different from 9 and 12° of freedom methods. Caudate nucleus volumes were expected to be associated with measures of processing speed and mental flexibility, but the strength of the association based on transformation approach was unknown. MRI brain scans from individuals with Parkinson's disease (n = 40) and age-matched controls (n = 40) were transformed using 6, 9, and 12° of freedom to an average brain template. Correlations controlling for total intracranial volume assessed expected structural-behavioral associations. Volumetric: Raw 9 and 12° transformed volumes were significantly larger than native and 6° volumes. Only 9 and 12° volumes revealed group differences with PD less than controls. Intracranial volume considerations were essential for native and 6° between group comparisons. Structural-Behavioral: The 9 and 12° caudate nucleus volume transformations revealed the expected brain-behavioral associations. Linear registration techniques alter volumetric and cognitive-structure associations. The study highlights the need to communicate transformation approach and group intracranial volume considerations.
脑磁共振成像(MRI)配准会改变结构的方向、大小和/或形状。为了确定线性配准方法(图像变换至6、9和12个自由度)是否会改变结构体积和认知关联,我们研究了被诊断为帕金森病(PD)的个体以及人口统计学匹配的非PD同龄人中尾状核的变换改变。来自原始图像和六种变换的体积预计与9和12自由度方法的体积有显著差异。尾状核体积预计与处理速度和心理灵活性的测量指标相关,但基于变换方法的关联强度尚不清楚。对帕金森病患者(n = 40)和年龄匹配的对照组(n = 40)的MRI脑部扫描进行6、9和12自由度的变换,使其与平均脑模板匹配。控制总颅内体积的相关性评估了预期的结构-行为关联。体积方面:原始的9和12自由度变换后的体积显著大于原始体积和6自由度变换后的体积。只有9和12自由度变换后的体积显示出组间差异,PD组小于对照组。在组间比较中,考虑颅内体积对于原始体积和6自由度变换后的体积至关重要。结构-行为方面:9和12自由度的尾状核体积变换显示出预期的脑-行为关联。线性配准技术会改变体积和认知-结构关联。该研究强调了交流变换方法和组间颅内体积考虑因素的必要性。