基于小波稀疏树形结构的MRI图像重建研究

MRI image reconstruction based on sparse wavelet-tree structure

核磁共振成像(MRI)作为现代医学科技手段一直限制于成像速度和易造成轮廓模糊的问题。而树形结构方法相比于传统K空间方法能进一步减少采样数据,加快处理速度和提高重建效果。文中提出了一种快速的CS-MRI凸优化算法,将全变分、小波变换和稀疏树形结构约束一并加入了模型中。通过引入增广拉格朗日乘子,用迭代更新中间变量的方法求解原问题。使原有的复杂原问题被分解为三个易迭代解决的子问题。多次试验证明这种新的CS-MRI算法相比于现有树形算法或重建图形算法有着更好的重建结果和更优越的信噪比。

Magnetic resonance imaging（ MRI） is an essential medical imaging tool burdened by an inherently slow data acquisition process but cause the fuzzy problem easily. According to structured sparsity theory,the measurements can be further reduced for tree-sparse data instead of for standard K-sparse data. This paper proposes a fast convex optimization algorithm to improve CS-MRI. Total variation,wavelet sparsity and tree sparsity are added into models. The augmented Lagrange multiplier is used to solve the original problem by the updating intermediate variable method. The original complex problem is decomposed into three simpler sub-problems and each of the sub-problems is solved with an iterative scheme.Numerous experiments show that the proposed algorithm outperforms the state-of-the-art CS-MRI algorithms and obtains better reconstruction results on real MR images than general tree based solvers or algorithms.