Optimal design of a blended diffusion/relaxometry experiment

Abstract:

Performing both diffusion (dMRI) and relaxometry MRI within a single clinical scan protocol is highly desirable as both modalities provide complementary information about the microstructure of brain tissues. Combining dMRI and relaxometry in a blended imaging protocol holds the potential to overcome limitations, e.g. in acquisition time or biomarker specificity. We propose a blended dMRI/relaxometry imaging protocol together with an optimal experiment design, based on the Cramér-Rao Lower Bound (CRLB), which allows increased precision of the model parameters of interest in a given acquisition time. To this end, we adopt the pulsed-gradient spin-echo sequence and a diffusion time-dependent model.

We proposed an optimized acquisition design for a blended dMRI/relaxometry protocol and compared it to an equidistant framework, based on a CRLB analysis and simulation experiments. The results show a gain in the precision of the estimates of the model parameters of interest in a given acquisition time, which contributes to the improvement of the trade-off between acquisition time and precision.

• Google Scholar
• BibTeX