文献速递妇产影像技术与安全MR

不同后处理软件生成的ADC图中绝对表观扩散系数(ADC)值的比较:子宫内膜癌中的可重复性

Comparison of Absolute Apparent Diffusion Coefficient (ADC) Values in ADC Maps Generated Across Different Postprocessing Software: Reproducibility in Endometrial Carcinoma.
PubMed 文献 · AJR Am J Roentgenol · 2017-09-27

本研究用1.5-T MRI扫描仪采集三个b值的扩散加权图像,将由扫描仪内置软件自动生成的ADC图与在后处理软件上另行生成的ADC图基于ROI进行比较,采用配对t检验、Bland-Altman图、山形图和Passing-Bablok回归图分析。结果显示,使用相同的基线trace扩散加权图像生成ADC图时,不同后处理软件所得平均ADC值之间存在统计学显著差异。作者认为,若要将ADC值作为组织学定性的定量截断值,则必须在各处理软件包之间标准化后处理算法,尤其考虑到厂商中立归档的实施。

要点

  • 同一基线trace扩散加权图像下,不同后处理软件所得平均ADC值存在统计学显著差异。
  • 比较方法包括配对t检验、Bland-Altman图、山形图和Passing-Bablok回归图。
  • 若将ADC值作为组织学定性的定量截断值,需在各软件包间标准化后处理算法。

核心结论

  • 使用相同的基线trace扩散加权图像生成ADC图时,不同后处理软件所得平均ADC值之间存在统计学显著差异。
  • 作者认为,若将ADC值作为组织学定性的定量截断值,则必须在各处理软件包之间标准化后处理算法,尤其考虑到厂商中立归档的实施。

研究设计与样本

  • 本研究为影像后处理软件比较研究,使用1.5-T MRI扫描仪采集三个b值的扩散加权图像并获得trace图像。
  • ADC图由扫描仪厂商提供的内嵌软件在图像生成时自动生成,同时也在后处理软件上另行生成。
  • 以ROI为基础,采用配对t检验、Bland-Altman图、山形图和Passing-Bablok回归图对ADC图进行比较。

局限性

  • 原文未报告局限性

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVE: Apparent diffusion coefficient (ADC) maps are usually generated by builtin software provided by the MRI scanner vendors; however, various open-source postprocessing software packages are available for image manipulation and parametric map generation. The purpose of this study is to establish the reproducibility of absolute ADC values obtained using different postprocessing software programs.

MATERIALS AND METHODS: DW images with three b values were obtained with a 1.5-T MRI scanner, and the trace images were obtained. ADC maps were automatically generated by the in-line software provided by the vendor during image generation and were also separately generated on postprocessing software. These ADC maps were compared on the basis of ROIs using paired t test, Bland-Altman plot, mountain plot, and Passing-Bablok regression plot.

RESULTS: There was a statistically significant difference in the mean ADC values obtained from the different postprocessing software programs when the same baseline trace DW images were used for the ADC map generation.

CONCLUSION: For using ADC values as a quantitative cutoff for histologic characterization of tissues, standardization of the postprocessing algorithm is essential across processing software packages, especially in view of the implementation of vendor-neutral archiving.

来源:AJR Am J Roentgenol,查看原文。版权归原作者所有。

引用

  1. Ghosh A, Singh T, Singla V, et al. Comparison of Absolute Apparent Diffusion Coefficient (ADC) Values in ADC Maps Generated Across Different Postprocessing Software: Reproducibility in Endometrial Carcinoma. AJR Am J Roentgenol. 2017. PMID: 28952809. doi:10.2214/AJR.17.18002

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。