OBJECTIVES To systematically test the reproducibility of DKI technique in normal liver and report a complete set of DKI measurement data. MATERIALS AND METHODS Thirty-two healthy volunteers were examined with liver DKI twice on the GE 3.0 T MRI scanner and reviewed by three professional experts. DKI-derived parameters fractional anisotropy of kurtosis (FAk), mean diffusivity (Md), axial diffusivity (Da), radial diffusivity (Dr), mean kurtosis (Mk), axial kurtosis (Ka), and radial kurtosis (Kr) in eight segments divided by Couinaud octagonal method were collected. Inter-class correlation coefficient (ICC) was used to assess the agreement between three experts. For each expert, the reproducibility of twice scans was evaluated by Bland-Altman method. Multivariate analysis of variance was to explore the regional distribution characteristics of DKI-derived parameters, and showed with box-plot graph. RESULTS Using ICC analysis, except for FAk (ICC 0.312, 0.307), other DKI metric values showed high reproducibility (0.716 < ICC < 0.907) between three experts for each of two DKI measurements. With Bland-Altman method, liver segment 5 (S5) showed the best reproducibility between two DKI measurement, and the reproducibility of segment 4 (S4) was the worst. The reproducibility of the right lobe was significantly higher than the left lobe. The values of diffusion metrics (Md, Da, and Dr) and kurtosis metrics (Mk, Ka, and Kr) existed significantly difference between the right and left hepatic lobes. CONCLUSION DKI has shown excellent reproducibility in liver imaging. The range of values for multiple DKI parameters, derived from the normal liver, was reported, and may provide data reference for further clinical DKI applications. Additionally, DKI technique is a non-invasive method to reflect the perfusion or structural differences between the left and right hepatic lobes from the molecular level.

中文翻译：

---

肝脏中的扩散峰度成像：健康志愿者的初步重现性研究。

目的系统地测试DKI技术在正常肝脏中的重现性，并报告一套完整的DKI测量数据。材料与方法 32 名健康志愿者在 GE 3.0 T MRI 扫描仪上用肝脏 DKI 进行了两次检查，并由 3 名专业专家审查。DKI 派生参数分数各向异性峰态 (FAk)、平均扩散率 (Md)、轴向扩散率 (Da)、径向扩散率 (Dr)、平均峰态 (Mk)、轴向峰态 (Ka) 和径向峰态 (Kr) 八收集通过 Couinaud 八角法分割的片段。类间相关系数（ICC）用于评估三位专家之间的一致性。对于每位专家，通过 Bland-Altman 方法评估两次扫描的再现性。多元方差分析是探索DKI衍生参数的区域分布特征，并用箱线图表示。结果 使用 ICC 分析，除 FAk (ICC 0.312, 0.307) 外，其他 DKI 度量值在三位专家之间显示出高重现性 (0.716 < ICC < 0.907)，分别用于两次 DKI 测量。Bland-Altman方法中，肝段5（S5）在两次DKI测量中的重现性最好，段4（S4）的重现性最差。右叶的再现性明显高于左叶。左右肝叶扩散指标（Md、Da、Dr）和峰度指标（Mk、Ka、Kr）存在显着差异。结论 DKI 在肝脏成像中显示出极好的重现性。报告了来自正常肝脏的多个 DKI 参数的值范围，可为进一步的临床 DKI 应用提供数据参考。此外，DKI技术是一种从分子水平反映左右肝叶灌注或结构差异的非侵入性方法。