|
KMID : 0372919980190060585
|
|
Journal of Biomedical Engineering Research
1998 Volume.19 No. 6 p.585 ~ p.593
|
|
|
Using Bayesian Approaches to Reduce Truncation Artifact in Magnetic Resonance Imaging
|
|
Lee Soo-Jin
|
|
|
Abstract
|
|
|
|
In Fourier magnetic resonance imaging (MRI), the number of phase encoded signals is often reduced to minimize the duration of the studies and maintain adequate signal-to-noise ratio. However, this results in the well-known truncation artifact, whose effect manifests itself as blurring and ringing in the image domain. In this paper, we propose a new regularization method in the context of a Bayesian framework to reduce truncation artifact. Since the truncation artifact appears in t도 phase direction only, the use of conventional piecewise-smoothness constraints with symmetric neighbors may result in the loss of small details and soft edge structures in the read direction. Here, we propose more elaborate forms of constraints than the conventional piecewise-smoothness constraints, which can capture actual spatial information about the MR images. Our experimental results indicate that the proposed method not only reduces the truncation artifact, but also improves tissue regularity and boundary definition without oversmoothing soft edge regions.
|
|
|
KEYWORD
|
|
|
Magnetic Resonance Imaging, Truncation Artifact, Bayesian Methods, Gibbs Distribution, Deterministic Annealing
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|