The effect of preprocessing methods for&nbsp; <sup>18</sup> F-FDG PET/CT images from different institutions in distinguishing lung adenocarcinoma <em>EGFR</em> mutation status

SUN Dazhen; LIU Qiufang; LI Nan; HUANG Gang; WANG Lisheng; SONG Shaoli

doi:10.19732/j.cnki.2096-6210.2021.06.006

您当前的位置：

首页 >

文章列表页 >

The effect of preprocessing methods for ¹⁸ F-FDG PET/CT images from different institutions in distinguishing lung adenocarcinoma EGFR mutation status

更新时间：2025-12-15

- The effect of preprocessing methods for ¹⁸ F-FDG PET/CT images from different institutions in distinguishing lung adenocarcinoma EGFR mutation status
- Vol. 30, Issue 6, Pages: 466-471(2021)
- 作者机构：
  
  1. 上海市分子影像学重点实验室，上海健康医学院,上海,201318
  2. 上海交通大学自动化系,上海,200210
  3. 复旦大学附属肿瘤医院核医学科，复旦大学上海医学院肿瘤学系,上海,200032
- 作者简介：
- 基金信息：
- DOI：10.19732/j.cnki.2096-6210.2021.06.006
  CLC：
- Published Online：28 December 2021，
  
  Published：28 December 2021
- 稿件说明：
移动端阅览
孙大桢,刘秋芳,李楠,等. 不同单位 ¹⁸ F-FDG PET/CT影像预处理对鉴别肺腺癌EGFR突变状态的影响[J]. 肿瘤影像学, 2021, 30(6): 466-471 https://doi.

org/10.19732/j.cnki.2096-6210.2021.06.006
孙大桢,刘秋芳,李楠,等. 不同单位 ¹⁸ F-FDG PET/CT影像预处理对鉴别肺腺癌EGFR突变状态的影响[J]. 肿瘤影像学, 2021, 30(6): 466-471 https://doi. DOI： 10.19732/j.cnki.2096-6210.2021.06.006.

org/10.19732/j.cnki.2096-6210.2021.06.006 DOI：

摘要

目的：

对来自两个不同中心的

F-FDG正电子发射体层成像（positron emission tomography，PET）/计算机体层成像（computed tomography，CT）影像进行影像组学分析，以预测肺腺癌表皮生长因子受体（epidermal growth factor receptor，

EGFR

）突变状态为任务，探究不同的多中心数据处理方式对特征稳定性和预测能力的影响，并寻找对多中心影像数据合适的处理方式。

方法：

对148例来自两个独立中心的非小细胞肺癌患者的PET/CT影像进行回顾性分析，这些影像采用不同的扫描设备，CT断层间隔分别为3和5 mm。在PET/CT影像上勾画感兴趣区（region of interest，ROI），对CT影像分别采用灰度范围归一化、降采样、插值升采样进行预处理后，提取、筛选影像组学特征，建立logistic回归（logistic regression，LR）模型。特征的稳定性通过统计学显著性检验的P值衡量。采用受试者工作特征（receiver operating characteristic，ROC）曲线评价模型效能，用来评估特征对EGFR突变状态的预测能力。

结果：

采用的3种图像预处理方法都能提升影像组学特征的稳定性，使用灰度范围归一化和降采样组合的方法效果最好，相比于基线能够多保留30.8%的特征，其对应特征的预测能力也最佳，构建的LR分类器在训练和测试集上曲线下面积（area under curve，AUC）分别为0.862和0.716。

结论：

灰度范围归一化配合降采样是有效的多中心影像预处理手段，能够提高影像组学特征的稳定性和预测效能。

Abstract

Objective:

To identify epidermal growth factor receptor (

EGFR

) mutation status lung adenocarcinoma patients using

F-FDG positron emission to

mography (PET)/computed tomography (CT) images from two different institutions

and to find appropriate imaging preprocessing methods to handle CT images of different slice thickness.

Methods:

F-FDG PET/CT images of 148 patients were retrospectively studied

which were scanned in two hospitals with different scanners and CT slice thickness (3 and 5 mm

respectively). The tumor region of interest (ROI) was manually delineated. Gray scale unifying

downsampling and upsampling with interpolation were used to pre-processing CT images. After the extraction and selection of radiomic features

logistic regression (LR) models were constructed to predict

EGFR

mutation status. Feature stability was evaluated through statistical tests. LR models were evaluated using the receiver operating characteristic (ROC) curve and served as an indicator of feature predicting power.

Results:

All adopted methods improves feature stability. The combination of gray scale unifying and downsampling performed best

resulting in 30.8% more robust features compared to baseline

and the corresponding LR model achieves area under curve (AUC) scores 0.862 in training and 0.716 in test cohort.

Conclusion:

The combination of gray scale unifying and downsampling is an effective approach for multi-center radiomic analysis

improving both feature stability and predictive power.

关键词

Keywords

references

Views

1835

下载量

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

The value of ¹⁸ F-FDG PET/CT radiomic features for predicting EGFR gene mutation status in patients with lung adenocarcinoma

Research status, development and clinical applications of PET/CT radiomics

Research progress of PET/CT radiomics based on artificial intelligence in clinical tumor diagnosis and treatment

The predictive value of metabolic parameters of ¹⁸ F-FDG PET/CT in patients with de novo metastatic nasopharyngeal carcinoma

Value of ¹⁸F-FDG PET/CT metabolic parameters in predicting lymph node metastasis in non-small cell lung cancer

Related Author

JIANG Xuemei

SHEN Jiaqi

PENG Xiaojuan

LI Peiqi

TAN Xiaofei

DANG Jun

YE Zhenyan

YOU Jinhui

Related Institution

Department of Nuclear Medicine, Sichuan Cancer Hospital and Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China

Medical Imaging School, North Sichuan Medical College, Nanchong

Shanghai Engineering Research Center of Molecular Imaging Probes

Key Laboratory of Nuclear Physics and Ion-beam Application, Fudan University

Center for Biomedical Imaging, Fudan University

AI问答

⁰

The effect of preprocessing methods for 18 F-FDG PET/CT images from different institutions in distinguishing lung adenocarcinoma EGFR mutation status

DOI：10.19732/j.cnki.2096-6210.2021.06.006

摘要

Abstract

关键词

Keywords

references

The effect of preprocessing methods for ¹⁸ F-FDG PET/CT images from different institutions in distinguishing lung adenocarcinoma EGFR mutation status