肝脏质子密度脂肪分数和R2^*测量ROI标准化方法探讨

ROI standardization method of liver proton density fat fraction and R2^*measurements

目的探讨化学位移编码(CSE)MRI基于ROI的肝脏质子密度脂肪分数(PDFF)和R2^*测量的标准化方法。方法回顾性分析79例腹部CSE数据,A组51例行常规腹部MRI受试者,男性25例,女性26例,平均年龄50.1岁。B组28例怀疑存在肝脏铁负荷患者,男性15例,女性13例,平均年龄47.2岁。三位评定者使用先前报道的ROI采样策略测量肝脏PDFF和R2^*。使用Bland-Altman分析评估评定者内部和评定者间测量PDFF和R2^*的一致性。结果在每个Couinaud肝段放置最大ROI,两组中PDFF和R2^*测量的一致性界限(LoA)均最窄。A、B组评定者间PDFF测量一致性LoA分别为±0.8%、±1.7%,评定者内部一致性LoA分别为±0.5%、±0.9%。A、B组评定者间R2^*测量一致性LoA分别±3.0/s、±17.9/s,评定者内部一致性LoA分别为±2.6/s、±14.6/s。在每个Couinaud肝段放置最大ROI分析相对耗时,每例需要(125.3±9.7)s。结论为提高肝脏PDFF和R2^*测量的可重复性,应使用多个大的ROI对尽可能多的肝脏区域进行采样。

Objective To investigate the standardized method of chemical shift-coding(CSE)MRI based on ROI-based liver proton density fat fraction(PDFF)and R2^*measurements.Methods We retrospectively analyzed 79 CSE data.Group A was 51 patients with conventional abdominal MRI(25 males,26 females;mean age 50.1 years),and group B was 28 patients with suspected liver iron load(15 males,13 females;mean age 47.2 years).Three raters used the previously reported ROI sampling strategy to measure liver PDFF and R2^*.Inter-and intra-assessor agreement of liver PDFF and R2^*were evaluated using Bland-Altman analysis.Results The maximum ROI was placed in each Couinaud segment,and the limits of agreement(LoA)of the liver PDFF and R2^*measurements were the narrowest in both groups.For PDFF measurements,LoA of inter-assessor agreement for groups A and B was±0.8%and±1.7%,respectively,and LoA of intra-assessor agreement for groups A and B was±0.5%and±0.9%,respectively.For R2^*measurement,LoA of inter-assessor agreement for groups A and B was±3.0/s and±17.9/s,respectively,and LoA of intra-assessor agreement for groups A and B was±2.6/s and±14.6/s,respectively.The maximum ROI analysis placed in each Couinaud segment was relatively time consuming,requiring 125.3±9.7 seconds per case.Conclusion To improve the repeatability of liver PDFF and R2^*measurements,multiple large ROIs should be used to sample as many liver regions as possible.