Bernard and Irene Schwartz Center for Biomedical Imaging, New York University School of Medicine, New York, New York, USA.
Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA.
J Magn Reson Imaging. 2021 Apr;53(4):1130-1139. doi: 10.1002/jmri.27421. Epub 2020 Nov 15.
3D-T mapping is useful to quantify various neurologic disorders, but data are currently time-consuming to acquire.
To compare the performance of five compressed sensing (CS) algorithms-spatiotemporal finite differences (STFD), exponential dictionary (EXP), 3D-wavelet transform (WAV), low-rank (LOW) and low-rank plus sparse model with spatial finite differences (L + S SFD)-for 3D-T mapping of the human brain with acceleration factors (AFs) of 2, 5, and 10.
Retrospective.
Eight healthy volunteers underwent T imaging of the whole brain.
FIELD STRENGTH/SEQUENCE: The sequence was fully sampled 3D Cartesian ultrafast gradient echo sequence with a customized T preparation module on a clinical 3T scanner.
The fully sampled data was undersampled by factors of 2, 5, and 10 and reconstructed with the five CS algorithms. Image reconstruction quality was evaluated and compared to the SENSE reconstruction of the fully sampled data (reference) and T estimation errors were assessed as a function of AF.
Normalized root mean squared errors (nRMSE) and median normalized absolute deviation (MNAD) errors were calculated to compare image reconstruction errors and T estimation errors, respectively. Linear regression plots, Bland-Altman plots, and Pearson correlation coefficients (CC) are shown.
For image reconstruction quality, at AF = 2, EXP transforms had the lowest mRMSE (1.56%). At higher AF values, STFD performed better, with the smallest errors (3.16% at AF = 5, 4.32% at AF = 10). For whole-brain quantitative T mapping, at AF = 2, EXP performed best (MNAD error = 1.62%). At higher AF values (AF = 5, 10), the STFD technique had the least errors (2.96% at AF = 5, 4.24% at AF = 10) and the smallest variance from the reference T estimates.
This study demonstrates the use of different CS algorithms that may be useful in reducing the scan time required to perform volumetric T mapping of the brain.
3D-T 映射对于量化各种神经疾病很有用，但目前获取数据的时间很长。
比较五种压缩感知（CS）算法（时空有限差分（STFD）、指数字典（EXP）、3D 小波变换（WAV）、低秩（LOW）和带空间有限差分的低秩稀疏模型（L+S SFD））在加速度因子（AF）为 2、5 和 10 时对人脑 3D-T 映射的性能。
回顾性。
8 名健康志愿者接受了全脑 T 成像。
磁场强度/序列：序列为完全采样的 3D 笛卡尔超快梯度回波序列，在临床 3T 扫描仪上使用定制的 T 准备模块。
对完全采样的数据进行了 2、5 和 10 的欠采样，并使用五种 CS 算法进行了重建。评估图像重建质量并与完全采样数据的 SENSE 重建（参考）进行比较，并评估 T 估计误差随 AF 的变化。
计算归一化均方根误差（nRMSE）和中位数归一化绝对偏差（MNAD）误差，分别比较图像重建误差和 T 估计误差。显示线性回归图、Bland-Altman 图和 Pearson 相关系数（CC）。
对于图像重建质量，在 AF = 2 时，EXP 变换具有最低的 mRMSE（1.56%）。在更高的 AF 值下，STFD 表现更好，误差最小（AF = 5 时为 3.16%，AF = 10 时为 4.32%）。对于整个大脑的定量 T 映射，在 AF = 2 时，EXP 表现最佳（MNAD 误差= 1.62%）。在更高的 AF 值（AF = 5、10）下，STFD 技术的误差最小（AF = 5 时为 2.96%，AF = 10 时为 4.24%），并且与参考 T 估计值的偏差最小。
本研究证明了不同 CS 算法的使用可能有助于减少进行大脑容积 T 映射所需的扫描时间。
2。
1。
