Potential of whole-liver MRI histogram analysis of remaining liver parenchyma to predict recurrence of colorectal liver metastases

LIANG Heyue; YANG Zhaoxia; FU Caixia; ZENG Mengsu; RAO Shengxiang

您当前的位置：

首页 >

文章列表页 >

Potential of whole-liver MRI histogram analysis of remaining liver parenchyma to predict recurrence of colorectal liver metastases

更新时间：2025-12-15

- Potential of whole-liver MRI histogram analysis of remaining liver parenchyma to predict recurrence of colorectal liver metastases
- Vol. 26, Issue 1, Pages: 18-24(2017)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：
  CLC：
- Published Online：13 April 2017，
  
  Published：13 April 2017
- 稿件说明：
移动端阅览
梁和月,阳朝霞,傅彩霞,等. 全肝MRI直方图分析检测结直肠癌肝转移复发的价值[J]. 肿瘤影像学, 2017, 26(1): 18-24
DOI：

梁和月,阳朝霞,傅彩霞,等. 全肝MRI直方图分析检测结直肠癌肝转移复发的价值[J]. 肿瘤影像学, 2017, 26(1): 18-24 DOI：

摘要

目的：

探讨治疗前MRI直方图分析在预测结直肠癌肝转移复发中的价值。

方法：

选取初次诊断为结直肠癌同步肝转移的50例患者，均接受包括扩散加权成像(diffusion weighted imaging，DWI)(b=0、500 s/mm

)在内的常规腹部增强MRI扫描。对表观扩散系数(apparent diffusion coefficient，ADC)和门脉期强化率图像进行直方图分析，获得包括均值、标准差、中位数、偏斜度、峰度及百分位数(1st、10th、50th、90th、99th百分位数)在内的参数。对所得直方图参数进行统计分析，比较肝转移复发(

=20)与无复发(

=30)组患者(随访间隔在6个月内)治疗前MRI直方图的差异。

结果：

结直肠癌肝转移患者门静脉期强化率直方图分析显示，复发组标准差和99th百分位数高于无复发组，差异有统计学意义(

=0.025，0.024)，曲线下面积均为0.68。而ADC直方图分析显示，均值、标准差、中位数、偏斜度、峰度和百分位数在复发组与无复发组之间均无统计学差异(

0.05)。

结论：

MRI门静脉期强化率直方图分析对预测结直肠癌肝转移患者的短期疗效具有潜在价值。

Abstract

Objective:

To explore whether magnetic resonance imaging (MRI) histogram analysis of apparently diseasefree liver parenchyma can help predict the recurrence of colorectal liver metastases.

Methods:

Standard enhanced MRI including diffusion weighted imaging (b=0

500 s/mm

) were performed at the first stage of diagnosis in 50 patients with colorectal liver metastases. Histograms were performed for apparent diffusion coefﬁcient (ADC) maps and enhanced rate image of portal

venous phase (PVP)

thereafter the mean

standard deviation

median

skewness

kurtosis and Nth percentiles (1st

10th

50th

90th

99th) were generated. Quantitative histogram parameters were compared between the patients with recurrence (

=20) and without recurrence (

=30) after a short period (about 6 months) of chemotherapy with or without surgery. Receiver operating characteristic (ROC) analyses was further analyzed for the significant parameters.

Results:

The mean

standard deviation

median

skewness

kurtosis and percentiles of ADC maps had no significant difference between the patients with and without hepatic recurrence (

0.05) after a short period of chemotherapy. The standard deviation and 99th percentile of enhanced rate image of portal venous phase were significantly higher in the patients with recurrence compared with those without recurrence (

=0.025

0.024)

and the areas under curve (AUCs) were both 0.68.

Conclusion:

Whole-liver MRI histogram analysis of apparently disease-free liver parenchyma can help predict hepatic recurrence in the patients with colorectal liver metastases.

关键词

Keywords

references

Views

1509

下载量

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

The value of multimodal imaging combined with CT, high-resolution MRI, and DWI in the diagnosis of postoperative anastomotic recurrence in rectal cancer

Association between the characteristics of conventional ultrasound combined with CEUS and the recurrence or metastasis of postoperative breast cancer

Advances in fatty acid metabolism in the tumor microenvironment of colorectal cancer

The value of color doppler ultrasound in the diagnosis of liver metastasis of gastrointestinal and extragastrointestinal malignant stromal tumor

The value of contrast-enhanced CT in assessing microsatellite instability with colorectal cancer

Related Author

Qingsong MA

Lulu LUO

Hongming LUO

Zejiang ZHANG

Lu LIU

LI Cuiying

PING Jieyi

DU Liwen

Related Institution

Department of Radiology, Ziyang Central Hospital

Department of Radiology, Shanghai Ninth People&rsquo

Department of Medical Imaging, Huadong Hosptial, Fudan University

Department of Radiology, Guangdong Provincial People&rsquo

CAS Key Laboratory of Molecular Imaging, Institute of Automation, Chinese Academy of Sciences

AI问答

⁰