ESTIMATION OF DIFFUSION MAGNETIC RESONANCE IMAGING SCALAR MEASURES FROM ARBITRARY Q-SPACE TRAJECTORIES
Publication Date: 2015-Mar-26
The IP.com Prior Art Database
The present invention provides a technique to estimate classical single pulsed field gradient (SPFG) diffusion measures from non-SPFG acquisitions. This saves time of acquiring SPFG measurements in cases when one is interested in both SPFG and non-SPFG measures. The technique includes simulating diffusion MRI signal from fitted diffusion and kurtosis tensors of SPFG datasets. The simulated signal from q-space trajectories is calculated by extending signal integral to the second-order cumulative term. The obtained signal is fed as inputs to a regression algorithm and the known scalar measures are predicted as outputs.
The present invention relates generally to diffusion magnetic resonance imaging (diffusion MRI) and more particularly to a technique for estimating diffusion MRI scalar measures from arbitrary q-space trajectories.
Generally, in classical single pulsed field gradient (SPFG) q-space imaging in diffusion MRI, several non-SPFG protocols are introduced in the state of the art. The non-SPFG protocols include rotating field gradient, q space magic angle spinning, q-space trajectory imaging, and generalized gradient waveforms, among others.
A conventional technique of calculating scalar measures is approximation, particularly machine learning. The technique constructs a computational model using Monte Carlo simulations and machine learning in order to learn a mapping between features derived from diffusion-weighted magnetic resonance (DW MR) signals and ground truth microstructure parameters.
Another conventional technique relates to image quality transfer. The technique provides fine structural detail of medical images from high quality data sets acquired with long acquisition times or from bespoke devices and transfer the information to enhance lower quality data sets from standard acquisitions. The technique utilizes random forest regression to relate patches in the low-quality data set to voxel values in the high quality data set.
However, the above mentioned conventional techniques do not use q-space trajectory data to estimate classical (non-trajectory) measures.
It would be desirable to have a technique to estimate diffusion MRI scalar measures from arbitrary q-space trajectories.
BRIEF DESCRIPTION OF DRAWINGS
Figure 1 depicts mean tensor kurtosis of (a) original data, (b) calculated from q-space trajectories, where the calculated values are integrals along, typically, circles in q-space simulated from fitted diffusion and kurtosis tensors and (c) used q-space trajectories.
The present invention provides a technique to estimate classical single pulsed field gradient (SPFG) diffusion measures from non-SPFG acq...