(a)
(b)
(c)
(d)
(e)
(f)
Figure 4: (a) to (c) show that the k-space data obtained with the conventional partial-Fourier DTI sequence, corresponding to different numbers of overscans, may be affected by a truncation artifact (i.e., Type 1 artifact) in the presence of an eddy current-induced echo-shifting effect. (d) to (f) show that this truncation artifact can be avoided in partial-Fourier DTI data obtained with the proposed k-space energy-anchored pulse sequence, where the k-space coverage moves with the echo-shifting effect.