Recent Developments in Fast Kurtosis Imaging

Abstract

Diffusion kurtosis imaging (DKI) is an extension of the popular diffusion tensor imaging (DTI) technique. DKI takes into account leading deviations from Gaussian diffusion stemming from a number of effects related to the microarchitecture and compartmentalization in biological tissues. DKI therefore offers increased sensitivity to subtle microstructural alterations over conventional diffusion imaging such as DTI, as has been demonstrated in numerous reports. For this reason, interest in routine clinical application of DKI is growing rapidly. In an effort to facilitate more widespread use of DKI, recent work by our group has focused on developing experimentally fast and robust estimates of DKI metrics. A significant increase in speed is made possible by a reduction in data demand achieved through rigorous analysis of the relation between the DKI signal and the kurtosis tensor based metrics. The fast DKI methods therefore need only 13 or 19 images for DKI parameter estimation compared to more than 60 for the most modest DKI protocols applied today. Closed form solutions also ensure rapid calculation of most DKI metrics. Some parameters can even be reconstructed in real time, which may be valuable in the clinic. The fast techniques are based on conventional diffusion sequences and are therefore easily implemented on almost any clinical system, in contrast to a range of other recently proposed advanced diffusion techniques. In addition to its general applicability, this also ensures that any acceleration achieved in conventional DKI through sequence or hardware optimization will also translate directly to fast DKI acquisitions. In this review, we recapitulate the theoretical basis for the fast kurtosis techniques and their relation to conventional DKI. We then discuss the currently available variants of the fast kurtosis techniques, their strengths and weaknesses, as well as their respective realms of application. These range from whole body applications to methods mostly suited for spinal cord or peripheral nerve, and analysis specific to brain white matter. Having covered these technical aspects, we proceed to review the fast kurtosis literature including validation studies, organ specific optimization studies and results from clinical applications.

Differential microstructural alterations in rat cerebral cortex in a model of chronic mild stress depression

“…We recently proposed a fast kurtosis method [ 45 , 47 ] enabling accurate estimation of mean, radial, and axial kurtosis on the basis of reduced data (fewer directions) compared to conventional diffusion kurtosis imaging [ 33 ]. A recent review paper provides an overview of the fast kurtosis methods ([ 48 ]. The kurtosis tensor method requires just two minutes of scan time [ 47 ] and the Matlab code to extract the kurtosis metrics is available at https://github.com/sunenj/Fast-diffusion-kurtosis-imaging-DKI .…”

Diffusion Kurtosis Imaging of Microstructural Changes in Gray Matter Nucleus in Parkinson Disease

“…One limitation of our study is that the spatial resolution is 1.875 × 1.875 × 4 mm 3 . This may yield imperfect reference values and is thus biased.…”

“…We hypothesized that there must be a connection inside the nuclei of the extrapyramidal system. Diffusion kurtosis imaging (DKI) offers improved sensitivity to tissue microstructure (3), which may be considered as an advantage over diffusion tensor imaging (DTI) (4). The basal ganglia and brainstem can be visualized in the DKI images (5).…”

Cerebral Damage after Carbon Monoxide Poisoning: A Longitudinal Diffusional Kurtosis Imaging Study

“…On the basis of previous studies, the b-value in WM should be higher than that in GM, where 2500$3000 s/mm 2 was found to be ideal in WM. 17,18 Therefore, the b-values of 0, 1000, and 3000 s/mm 2 were used in this scan.…”