Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
| Hi-index | 0.00 |
Robust field map estimation is important to many MRI applications, such as reconstruction with correction of susceptibility artifacts, MR-based temperature mapping, and waterfat separation. To enable in-vivo field map estimation with minimal scan times, multi-echo imaging sequences, which acquire multiple images in a single repetition, are gaining great interest. However, it has been observed that field map estimation becomes less reliable with multi-echo imaging sequences, especially at high field strengths and around challenging anatomies where good shimming cannot be obtained. In this paper, the field map estimation is shown to be a high-dimensional combinatorial optimization problem, which cannot be addressed by local greedy algorithms. This paper describes an effective approach based on message passing algorithm to globally approximate a solution with maximum a posterior (MAP) probability.