Canon Medical Systems Corporation, Otawara, Japan.
Canon Medical Systems Corporation, Otawara, Japan; Division of Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA; Department of Radiology, University of Wisconsin, Madison, WI, USA.
Magn Reson Imaging. 2022 Jan;85:57-63. doi: 10.1016/j.mri.2021.10.020. Epub 2021 Oct 19.
The effect of hepatic steatosis on the gradient-echo (GRE) based Modified Look-Locker Inversion Recovery (MOLLI) technique for T1 mapping has not been evaluated. The purpose of this study was to evaluate a GRE based MOLLI technique for hepatic T1 mapping and determine the relationship of T1 differences (ΔT1) on in-phase (IP) and out-of-phase (OP) to fat fraction (FF) measurement.
3 T MRI included MOLLI T1 mapping with TE = 1.3 (OP), 2.4 (IP), and 1.8 ms, and chemical-shift-encoded sequence with spectral modeling of fat to generate FF map as a reference. Bloch simulations and oil/water phantoms were used to characterize the response of the MOLLI T1 in various FF < 30% since MOLLI T1 estimation was erratic beyond this limit. Curve fit between ΔT1 and FF from simulation was applied to validate the phantom and the in-vivo results. Thirty-eight normal volunteers were included (16 women, Age 44 ± 12 years, BMI 27 ± 5.3 kg/m). MOLLI water images were reconstructed by the average of OP and IP images, and the T1 values on water images served as the reference for T1 bias calculation defined as the percent difference between OP, IP, TE = 1.8 ms and the referenced water T1. Linear regression was performed to correlate the FF quantified by the reference and MOLLI methods.
Phantom results were consistent with the Bloch simulations. The simulated relationship between FF (0-30%) and ΔT1 could be modeled precisely by a cubic equation with R2 = 1. In-vivo MOLLI ΔT1 and estimated FF were correlated to the reference FF (both R2 ≥ 0.96 and P < 0.001). TE = 1.8 ms demonstrated less T1 bias (-1.34%) compared to TE = OP (5.32%) or IP (-3.8%, both P < 0.001).
At 3 T, TE of 1.8 ms can be used to reduce the T1 bias and deliver consistent T1 values when FF is <30%.
尚未评估肝脂肪变性对基于梯度回波（GRE）的改良 Look-Locker 反转恢复（MOLLI）技术用于 T1 映射的影响。本研究的目的是评估基于 GRE 的 MOLLI 技术进行肝 T1 映射，并确定同相位（IP）和反相位（OP）上的 T1 差异（ΔT1）与脂肪分数（FF）测量值之间的关系。
在 3T MRI 中，MOLLI T1 映射的 TE 分别为 1.3（OP）、2.4（IP）和 1.8ms，并采用化学位移编码序列进行频谱建模，以生成 FF 图作为参考。采用 Bloch 模拟和油水体模来描述各种 FF <30%时 MOLLI T1 的响应，因为超出此限制后，MOLLI T1 估计会变得不稳定。对模拟中 ΔT1 与 FF 的曲线拟合应用于验证体模和体内结果。共纳入 38 名正常志愿者（16 名女性，年龄 44 ± 12 岁，BMI 27 ± 5.3kg/m2）。通过对 OP 和 IP 图像进行平均重建 MOLLI 水图像，并将水图像上的 T1 值作为 T1 偏差计算的参考，该偏差定义为 OP、IP、TE = 1.8ms 与参考水 T1 之间的百分比差异。采用线性回归来关联参考和 MOLLI 方法量化的 FF。
体模结果与 Bloch 模拟一致。FF（0-30%）与 ΔT1 之间的模拟关系可以通过 R2 = 1 的三次方程精确建模。体内 MOLLI ΔT1 和估计的 FF 与参考 FF 相关（均 R2 ≥ 0.96，P < 0.001）。与 TE = OP（5.32%）或 IP（-3.8%，均 P < 0.001）相比，TE = 1.8ms 时的 T1 偏差较小（-1.34%）。
在 3T 时，当 FF <30%时，可采用 1.8ms 的 TE 来降低 T1 偏差并获得一致的 T1 值。
