Systematic Comparison of DTI at 7T and 3T: Assessment of FA for Different Acquisition Protocols and SNR in Healthy Subjects

Introduction: Diffusion tensor imaging (DTI) has many promising clinical applications but is limited at ultra-high field. Higher field strengths increase SNR but are more prone to artifacts from B0 and B1 field inhomogeneity and shorter T2*. There are several studies exploring the feasibility of 7T DTI . In DTI applications, accurate measurement of fractional anisotropy (FA) is critical in assessment of white matter (WM) integrity and its microstructure. Recently, cross-site difference in bias level of DTI measurement was reported, in spite of that the measurement carried out by scanners and software were from the same vendor . Further, each group has developed particular acquisition methods and scan parameters. Thus, FA will have certain range of measured values method-to-method as well as site-to-site. In this study, we explored the effect of different scan parameters on FA values in aging brain at 7T and 3T. In the study, we considered the voxel shape, the number of diffusion-weighting (DW) gradients, and voxel volume as well as FA sampling methods. In addition, we investigated DTI SNR and FA over a range of SENSE factor (R) systematically. As a trial, we explored high resolution DTI at 7T and 3T for comparison.