Figure - PMC

Skip to main content

View full-text article in PMC

. Author manuscript; available in PMC: 2018 Aug 1.

Published in final edited form as: Magn Reson Med. 2017 May 28;78(2):419–428. doi: 10.1002/mrm.26762

FIG. 1 — An example of the proposed acquisition scheme. The “training” dataset $D_{1, T}$ is acquired at high temporal sampling rate but with limited k-space coverage (blue boxes). The sparse “imaging” dataset $D_{2}$ is acquired with extended k-space coverage (green cross signs). The acquisitions of $D_{1, T}$ and $D_{2}$ are interleaved in the whole dynamic process of the imaging object. Another “training” dataset $D_{1, f}$ is acquired with high spectral bandwidth but limited k-space coverage (red box) when the studied dynamic process is relatively stationary. Signal averaging can be used to improve SNR when acquiring $D_{1, f}$ (as reflected by the width of the red box).