Comparison of parameter optimization methods for quantitative susceptibility mapping.
Comparison of parameter optimization methods for quantitative susceptibility mapping.
复制标题
DOI：
10.1002/mrm.28435
复制
发表时间：
2021-01
影响因子：
3.3
通讯作者：
Tejos C
中科院分区：
文献类型：
--
作者：
Milovic C;Prieto C;Bilgic B;Uribe S;Acosta-Cabronero J;Irarrazaval P;Tejos C
定量敏感性映射通常通过最小化具有数据保真度和正则化项的函数来实现。加权参数控制这些项之间的平衡。需要技术来找到适当的平衡，以避免工件传播和细节丢失。在l曲线中寻找最大曲率点是一种流行的选择，尽管速度很慢，在使用变分惩罚时通常不可靠，并且倾向于产生过度正则化的结果。我们提出了两种替代方法来控制数据保真度和正则化项之间的平衡：1)在l曲线的log-log域中搜索拐点，2)比较QSM重构的频率成分。我们将这些方法与传统的l曲线和u曲线方法进行比较。与传统方法相比，我们的方法获得的预测参数与基于RMSE、HFEN和ssim的参数优化具有更好的相关性。与传统方法相比，拐点产生的过度正则化和误差更小。频率分析产生了更吸引人的视觉结果，尽管RMSE更大。我们的方法为QSM重构中的变分惩罚提供了一个鲁棒的参数优化框架。对于典型的QSM获取设置，基于l曲线的零曲率搜索产生了几乎最佳的结果。频率分析方法可以使用1.5-2.0校正因子将其作为独立方法应用于更宽范围的信噪比设置。这种方法还可能受益于快速搜索算法，如二进制搜索，以加快该过程。
Quantitative susceptibility mapping is usually performed by minimizing a functional with data fidelity and regularization terms. A weighting parameter controls the balance between these terms. There is a need for techniques to find the proper balance that avoids artifact propagation and loss of details. Finding the point of maximum curvature in the L-curve is a popular choice, although slow, often unreliable when using variational penalties, and tends to yield over-regularized results. We propose two alternative approaches to control the balance between the data fidelity and regularization terms: 1) searching for an inflection point in the log-log domain of the L-curve, and 2) comparing frequency components of QSM reconstructions. We compare these methods against the conventional L-curve and U-curve approaches. Our methods achieve predicted parameters that are better correlated with RMSE, HFEN and SSIM-based parameter optimizations than those obtained with traditional methods. The inflection point yields less over-regularization and lower errors than traditional alternatives. The frequency analysis yields more visually appealing results, although with larger RMSE. Our methods provide a robust parameter optimization framework for variational penalties in QSM reconstruction. The L-curve based zero-curvature search produced almost optimal results for typical QSM acquisition settings. The frequency analysis method may use a 1.5–2.0 correction factor to apply it as a standalone method for a wider range of SNR settings. This approach may also benefit from fast search algorithms such as the binary search to speed-up the process.
登录
查看更多内容
定量敏感性映射：2016年重建挑战的报告。
影响因子：
3.3
作者：
Langkammer C;Schweser F;Shmueli K;Kames C;Li X;Guo L;Milovic C;Kim J;Wei H;Bredies K;Buch S;Guo Y;Liu Z;Meineke J;Rauscher A;Marques JP;Bilgic B
通讯作者：
Bilgic B
影响因子：
3.3
作者：
Milovic C;Tejos C;Acosta-Cabronero J;Özbay PS;Schwesser F;Marques JP;Irarrazaval P;Bilgic B;Langkammer C
通讯作者：
Langkammer C
DOI：
10.1002/cmr.b.10083
发表时间：
2003-10-01
影响因子：
0.9
作者：
Salomir, R;De Senneville, BD;Moonen, CTW
通讯作者：
Moonen, CTW
影响因子：
3.4
作者：
Chen, Maomao;Su, Han;Luo, Jianwen
通讯作者：
Luo, Jianwen
影响因子：
4
作者：
Chamorro-Servent, Judit;Dubois, Remi;Coudiere, Yves
通讯作者：
Coudiere, Yves