Figure 8.

Evaluation of phase correction procedure: model‐based reconstruction with and without phase correction for dynamic frames of data acquired at b‐value of 500 s/mm². Reconstructions for each individual diffusion direction (Dir 1, Dir 2, and Dir 3) are shown for patient 11 and patient 1 in (a) and (b), respectively. Data are combined from 2 dynamics (dynamic 1 and 13 for patient 11, dynamic 1 and 25 for patient 1) that had significantly different phases in conjugate phase reconstructions. Without phase correction, the phase cancellation in each model‐based reconstruction leads to signal dropout and signal cancellations (bottom row in a and b). The proposed phase correction procedure preserved signal in each diffusion direction (top row in a and b). The magnitude average (Mag avg) of model‐based reconstructions from the 3 directions (Dir 1, Dir 2, and Dir 3) is shown in the right column