基于小波收缩减小磁共振图像截断伪影的方法

在磁共振成像（MRI）中，为了缩短成像时间并使图像保持较高的信噪比，往往只采集部分相位编码的磁共振信号进行图像重建，但在重建的图像上会出现截断伪影．在分析磁共振图像截断伪影特征的基础上，为了尽可能保留图像的细节特征，利用小波变换的多分辨率特性，对与截断伪影相关的细节子带进行小波收缩，并利用图像模板技术进一步消除图像背景区域的截断伪影和噪声．实验结果表明，该方法能有效地减小截断伪影，同时能更好地保持图像的细节特征．

Reduction of MRI Truncation Artifacts Based on Wavelet Shrinkage

In order to minimize the duration of magnetic resonance imaging and maintain adequate signal-to-noise ratio in Magnetic Resonance Imaging(MRI),a part of the phase-coded signals were often acquired to construct image.However,this results in truncation artifacts in the image.After analyzing the characteristics of the truncation artifacts in MRI,an algorithm based on wavelet shrinkage theory and the mask of image was proposed to reduce the truncation artifacts in magnetic resonance image.Wavelet shrinkage was performed to the sub-bands which result in the truncation artifacts only.In order to improve the image quality further,image mask was applied to remove the residue noise and truncation artifacts in the background of the image.Experimental results show that the proposed method can reduce the truncation artifacts efficiently and preserve the feature of the image as much as possible.