Multi-centre reproducibility of diffusion MRI parameters for clinical sequences in the brain

Abstract

The purpose of this work was to assess the reproducibility of diffusion imaging, and in particular the apparent diffusion coefficient (ADC), intra-voxel incoherent motion (IVIM) parameters and diffusion tensor imaging (DTI) parameters, across multiple centres using clinically available protocols with limited harmonization between sequences. An ice-water phantom and nine healthy volunteers were scanned across fives centres on eight scanners (four Siemens 1.5T, four Philips 3T). The mean ADC, IVIM parameters (diffusion coefficient D and perfusion fraction f) and DTI parameters (mean diffusivity MD and fractional anisotropy FA), were measured in grey matter, white matter and specific brain sub-regions. A mixed effect model was used to measure the intra- and inter-scanner coefficient of variation (CV) for each of the five parameters. ADC, D, MD and FA had a good intra- and inter-scanner reproducibility in both grey and white matter, with a CV ranging between 1% and 7.4%; mean 2.6%. Other brain regions also showed high levels of reproducibility except for small structures such as the choroid plexus. The IVIM parameter f had a higher intra-scanner CV of 8.4% and inter-scanner CV of 24.8%. No major difference in the inter-scanner CV for ADC, D, MD and FA was observed when analysing the 1.5T and 3T scanners separately. ADC, D, MD and FA all showed good intra-scanner reproducibility, with the inter-scanner reproducibility being comparable or faring slightly worse, suggesting that using data from multiple scanners does not have an adverse effect compared with using data from the same scanner. The IVIM parameter f had a poorer inter-scanner CV when scanners of different field strengths were combined, and the parameter was also affected by the scan acquisition resolution. This study shows that the majority of diffusion MRI derived parameters are robust across 1.5T and 3T scanners and suitable for use in multi-centre clinical studies and trials.

Keywords: MRI; brain; diffusion; multi-centre; reproducibility.

Figure 1

Segmentation of T₁-weighted image using FreeSurfer. T₁-weighted images were segmented using FreeSurfer in order to create masks defining 1, cerebellar cortex, 2, cerebellar white matter, 3, brain stem, 4, cerebral white matter, 5, basal ganglia (including the caudate nucleus, the putamen and the globus pallidus), 6, thalamus, 7, choroid plexus, and the optic chiasm (not shown).

Figure 2

Measuring FA in white matter areas. The segmented white matter mask (yellow) is shown overlaid on a high-resolution T₁-weighted image registered to standard MNI space. The ICBM-DTI-81 atlas (coloured areas) was used in order to measure the FA in areas defined as white matter according to the segmented mask in specific tracts: 1, genu of the corpus callosum; 2, body of the corpus callosum; 3, splenium of the corpus callosum; 4, superior longitudinal fasciculus; 5, cerebral peduncle; 6, sagittal stratum; 7, cingulum; 8, uncinate fasciculus.

Figure 3

Comparison of the ice–water phantom ADC images across scanners. Images of the phantom from each of the scanners using the same contrast range are shown. The tube filled with sucrose appears darker, while the five tubes filled with distilled water are evenly separated from each other and can be seen to be surrounded by ice–water. The ice–water was prepared by using either ice cubes (e.g. scanner A) or crushed ice (e.g. scanner C). The protocol for scanning the phantom was not adhered to for scanner F, with the image showing that ice–water was not surrounding all of the tubes at acquisition.

Figure 4

Box plots for DWI and DTI parameters across all scanners in grey matter (GM). ADC, D and MD are shown with the same range on the y-axis for direct comparison. A–H represent each scanner involved in the study, and the red data points represent individual subjects. ADC and MD had very similar values, while D had comparable but lower values. The boxplots confirm the higher values of f in scanner G.

Figure 5

Box plots for DWI and DTI parameters across all scanners in white matter (WM). ADC, D and MD are shown with the same range on the y-axis for direct comparison. A–H represent each scanner involved in the study, and the red data points represent individual subjects. ADC and MD had very similar values, while D had comparable but lower values. The boxplots confirm the higher values of f in scanner G.

Figure 6

Inter- and intra-scanner CV in volunteers. The graph shows the inter- and intra-scanner CV for all parameters in overall grey matter (GM) and overall white matter (WM), together with the mean values for each of the parameters. The inter-scanner CV was less than 4% for all parameters except for the perfusion fraction. The higher intra-scanner CV for FA in grey matter is likely to be due to an increased noise effect on FA values closer to zero, driving the intra-scanner variation up.

Figure 7

Inter-scanner CV by field strength. A comparison of the inter-scanner CV in overall grey matter (GM) and overall white matter (WM) for the model constructed using only 1.5T scanners, only 3T scanners and both field strengths is shown. The CV was less than 4% for all parameters except for the perfusion fraction.