Abstracts

Rationale: The aim of this study is to quantify the benefit of 7T imaging for assessing neuroanatomical changes. Recent developments in 7T MRI allow for the acquisition of isotropic whole brain MRI with superior contrast and spatial resolution to 3T MRI. We used 7T MRI to measure hippocampal volume and vertex-wise cortical thickness, and compared these assessments with conventional 3T imaging. Power analyses were used to quantify the benefit of high field MRI by estimating the minimum volume or thickness difference that could be identified using 7T and 3T imaging.Methods: Whole brain T1-weighted MRI was acquired at 7T and 3T (Siemens 7T whole body in pTx configuration; Siemens 3T Trio) in 9 subjects. The 7T data were acquired with a 8x2 16 channel transceiver, B1 shimmed for homogeneous mode imaging (Resonance Research Inc.). 3T data were acquired with a 32 channel head coil. The 3D 7T data used MP2RAGE with 4° and 5° flip angles, TR/TE 6s/3.16ms, GRAPPA 3, acquisition duration 9.5min, 750um isotropic. The 3T data used MPRAGE with 8° flip angle, TR/TE 1.56/3.2, GRAPPA 2, acquisition duration 5.1min, 1 mm isotropic. Freesurfer v 5.3 was used for subcortical segmentation and cortical surface modeling. The standard freesurfer processing stream was used for the 3T MRI. A previously published image processing protocol [1] was used for the 7T imaging data that allowed morphometric estimates to be assessed at the native resolution (0.75 mm isotropic). 7T and 3T across-subject variance estimates were used to calculate the minimum detectable effect size for hippocampal volume and vertex-wise cortical thickness using power analysis methods implemented in the R software package.Results: An example of 7T and 3T images of the same subject are shown in Fig. 1. Across-subject variance was lower for hippocampal volumes and average cortical thickness when assessed at 7T compared with 3T imaging. Power analyses demonstrate that 7T imaging may be used to detect smaller hippocampal volume or cortical thickness changes than 3T imaging in group studies with equivalent numbers of participants (Figure 2 left). Conversely, the same effect size could be detected at 7T using lower numbers of participants than 3T. For example a hippocampal volume difference of 400 mm3 could be detected using 30 subjects at 3T or less than 10 subjects at 7T. Although there is an overall improvement in sensitivity for cortical thickness measurement at 7T across the cortex, some cortical regions have higher variance at 7T such as the lateral temporal cortical regions (Figure 2 right).Conclusions: We have demonstrated that the higher resolution and contrast 7T imaging may be used to obtain quantitative morphometric estimates with lower variance than 3T imaging. Increased statistical power associated with 7T imaging will allow for the detection of subtle hippocampal sclerosis or focal cortical dysplasia, or the ability to identify neuroanatomical changes in smaller groups which will be beneficial for studying rare epilepsy syndromes. [1] Lusebrink et al, Neuroimage (2013) 70:122-31