Radiology, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Medical Physics, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Magn Reson Med. 2022 Dec;88(6):2662-2678. doi: 10.1002/mrm.29406. Epub 2022 Aug 15.
The purpose of this work was to obtain precise tri-exponential intravoxel incoherent motion (IVIM) quantification in the liver using 2D (b-value and first-order motion moment [M ]) IVIM-DWI acquisitions and region of interest (ROI)-based fitting techniques.
Diffusion MRI of the liver was performed in 10 healthy volunteers using three IVIM-DWI acquisitions: conventional monopolar, optimized monopolar, and optimized 2D (b-M ). For each acquisition, bi-exponential and tri-exponential full, segmented, and over-segmented ROI-based fitting and a newly proposed blood velocity SDdistribution (BVD) fitting technique were performed to obtain IVIM estimates in the right and left liver lobes. Fitting quality was evaluated using corrected Akaike information criterion. Precision metrics (test-retest repeatability, inter-reader reproducibility, and inter-lobar agreement) were evaluated using Bland-Altman analysis, repeatability/reproducibility coefficients (RPCs), and paired sample t-tests. Precision was compared across acquisitions and fitting methods.
High repeatability and reproducibility was observed in the estimations of the diffusion coefficient (D = [1.03 ± 0.11] × 10 mm /s; RPCs ≤ 1.34 × 10 mm /s), perfusion fractions (F = 3.19 ± 1.89% and F = 16.4 ± 2.07%; RPCs ≤ 2.51%), and blood velocity SDs (V = 1.44 ± 0.14 mm/s and V = 3.62 ± 0.13 mm/s; RPCs ≤ 0.41 mm/s) in the right liver lobe using the 2D (b-M ) acquisition in conjunction with BVD fitting. Using these methods, significantly larger (p < 0.01) estimates of D and F were observed in the left lobe in comparison to the right lobe, while estimates of V and V demonstrated high interlobar agreement (RPCs ≤ 0.45 mm/s).
The 2D (b-M ) IVIM-DWI data acquisition in conjunction with BVD fitting enables highly precise tri-exponential IVIM quantification in the right liver lobe.
本研究旨在使用二维(b 值和一阶运动矩 [M ])IVIM-DWI 采集和基于感兴趣区(ROI)的拟合技术,在肝脏中获得精确的三指数体素内不相干运动(IVIM)定量。
对 10 名健康志愿者进行肝脏扩散 MRI 检查,使用三种 IVIM-DWI 采集方法:常规单极、优化单极和优化二维(b-M )。对于每种采集方法,均进行双指数和三指数全、分段和过分段 ROI 拟合,以及新提出的血流速度标准差分布(BVD)拟合技术,以获得右肝和左肝的 IVIM 估计值。使用校正的赤池信息量准则评估拟合质量。使用 Bland-Altman 分析、重复性/再现性系数(RPCs)和配对样本 t 检验评估重复性、再现性和肝内一致性的精度指标。比较了不同采集方法和拟合方法之间的精度。
使用二维(b-M )采集和 BVD 拟合方法,在右肝的扩散系数(D = [1.03 ± 0.11] × 10 毫米 /秒;RPCs ≤ 1.34 × 10 毫米 /秒)、灌注分数(F = 3.19 ± 1.89%和 F = 16.4 ± 2.07%;RPCs ≤ 2.51%)和血流速度标准差(V = 1.44 ± 0.14 毫米/秒和 V = 3.62 ± 0.13 毫米/秒;RPCs ≤ 0.41 毫米/秒)的估计中,具有很高的重复性和再现性。使用这些方法,与右肝相比,左肝的 D 和 F 估计值显著增大(p < 0.01),而 V 和 V 的估计值具有很高的肝内一致性(RPCs ≤ 0.45 毫米/秒)。
二维(b-M )IVIM-DWI 数据采集与 BVD 拟合相结合,可在右肝实现高度精确的三指数 IVIM 定量。