Correlation Tensor Magnetic Resonance Imaging

Abstract

Diffusional Kurtosis Imaging (DKI) is a sensitive biomarker for microstructure in health and disease. However, DKI is not specific to any microstructural property since it may emerge from several different sources. Q-space trajectory encoding has been proposed for decoupling isotropic from anisotropic kurtosis. Still, this method assume that the system is comprised of multiple Gaussian diffusion components. Here, we develop a more general framework for resolving the underlying kurtosis sources. We introduce Correlation Tensor MRI (CTI) - an approach harnessing the versatility of double diffusion encoding (DDE) and capable of explicitly decoupling isotropic and anisotropic kurtosis components from intra-compartmental kurtosis effects arising from restricted diffusion. Additionally, CTI provides an index that is potentially sensitive to intra-compartmental kurtosis. The theoretical foundations of CTI, as well as the first proof-of-concept CTI ex vivo experiments in mouse brains at a field of 16.4T, are presented. We find that anisotropic and isotropic kurtosis can decouple microscopic anisotropy from substantial partial volume effects between tissue and free water. Our intra-compartmental kurtosis index exhibited positive values in both white and grey matter tissues. Simulations in different microenvironments show, however, that our current CTI protocol for estimating intra-compartmental kurtosis is limited by higher-order terms that were not considered in this study. CTI measurements were then extended to in vivo settings and used to map healthy rat brains at 9.4T. These in vivo CTI results were consistent with our ex vivo findings. Although future studies are still required to mitigate the higher-order effects on the intra-compartmental kurtosis index, results show that CTI's more general estimates of anisotropic and isotropic kurtosis contributions are already ripe for future studies.