Measuring water exchange on a preclinical MRI system using filter exchange and diffusion time dependent kurtosis imaging

Purpose: Filter exchange imaging (FEXI) and diffusion time (t)-dependent diffusion kurtosis imaging (DKI(t)) are both sensitive to water exchange between tissue compartments. The restrictive effects of tissue microstructure, however, introduce bias to the exchange rate obtained by these two methods, as their interpretation conventionally rely on the Karger model of barrier limited exchange between Gaussian compartments. Here, we investigated whether FEXI and DKI(t) can provide comparable exchange rates in ex vivo mouse brains. Theory and Methods: FEXI and DKI(t) data were acquired from ex vivo mouse brains on a preclinical MRI system. Phase cycling and negative slice prewinder gradients were used to minimize the interferences from imaging gradients. Results: In the corpus callosum, apparent exchange rate (AXR) from FEXI correlated with the exchange rate (the inverse of exchange time, 1/tau(ex)) from DKI(t) along the radial direction. In comparison, discrepancies between FEXI and DKI(t) were found in the cortex due to low filter efficiency and confounding effects from tissue microstructure. Conclusion: The results suggest that FEXI and DKI(t) are sensitive to the same exchange processes in white matter when separated from restrictive effects of microstructure. The complex microstructure in gray matter, with potential exchange among multiple compartments and confounding effects of microstructure, still pose a challenge for FEXI and DKI(t).