Determination of optimum combination of voxel size and b-value for brain diffusion tensor imaging

Abstract

Optimum combination of voxel size resolution and b-value for whole brain imaging has been determined. Data images were acquired using a 1.5T magnetic resonance imaging (MRI) system (GE Signa HDxt). Diffusion tensor imaging (DTI) scan was performed on phantom and a human volunteer. Six protocols which consist of various combination of voxel size and b-value were evaluated. Measurement of signal-to-noise ratio (SNR) and DTI parameter indices were carried out for both phantom and in-vivo studies. Due consideration was given to a combination of parameters yielding sufficient SNR with DTI values comparable to those obtained from previous reported studies. For the phantom study, SNR ≥ 20 was found in all of the protocols except for a combination of voxel size of 2.0 × 2.0 × 2.0 mm3 with b-value of 1200 s/mm2 (V2.0 B1200) and that of voxel size of 2.0 × 2.0 × 2.0 mm3 with b-value of 1000 s/mm2 (V2.0 B1000). For in-vivo study, all protocols presented SNR > 20. It was found that a combination of voxel size of 2.5 × 2.5 × 2.5 mm3 with b-value of 1000 s/mm2 (V2.5 B1000) and that of voxel size of 2.5 × 2.5 × 2.5 mm3 with b-value of 700 s/mm2 (V2.5 B700) displayed the most comparable ADC and FA values with references. In terms of anatomic coverage, V2.5 B700 was found better than V2.5 B1000 as it assures coverage of the whole brain. In conclusion, a combination of voxel size of 2.5 × 2.5 × 2.5 mm3 with b-value of 700 s/mm2 was considered as optimum parameters for brain DTI.