基于l1范数与全变分正则化的磁化率分布重建

Quantitative susceptibility mapping reconstruction based on l1 norm and total variation regularization

磁共振成像中相位图像包含丰富的组织磁化率变化信息,同时获取相位图像不需要额外的扫描时间。定量磁化率成像技术目前已经成为科学和临床研究中的一个热点问题,这种技术可以对组织内顺磁性物质进行定量分析,能够比常规的磁共振成像方法提供更多图像信息,有利于许多神经系统疾病和脑血管疾病的诊断。组织中的顺磁性物质会影响组织磁化率差异从而引起局部不均匀磁场,然而从局部相位信息重建组织磁化率分布是一个病态逆问题,目前仍然有许多问题亟待解决。提出一种基于l1范数与全变分正则化模型相结合的磁化率分布稀疏重建方法,仿真实验结果表明该方法可以有效获得高质量的定量磁化率分布图,提高了重建磁化率信息的准确性。

The phase image of magnetic resonance imaging （MRI） contains lots of tissue susceptibility variation information, while obtaining the phase image does not require additional scan time. Quantitative susceptibility mapping has become a hot issue in scientific and clinical research. This technique can realize the quantitative analysis of the paramagnetic in organization, and it can provide more image information than conventional MRI method, which is conducive to diagnosis of the eerebrovascular and nervous system diseases. The paramagnetic of the tissue can affect the differences of the tissue susceptibility and introduce about local inhomogeneous magnetic field consequently. However, the reconstruction of magnetic susceptibility distribution from local phase information is an ill-posed inverse problem. There are still many problems to be solved. This paper presents a new mathematics model of the susceptibility sparse reconstruction based on ll norm and total variation regularization method. The simulation experimental results show that this method can effectively obtain high quality images of quantitative susceptibility and improve the reconstruction accuracy of the structure information in the magnetic susceptibility distribution.