Comparative study on the utility of different radiomics feature selection methods in predicting survival of early-stage NSCLC patients

WANG Xinchao; CUI Caozhe; HU Yiyi; LI Xiaomeng; MENG Xiaxia; WU Zhifang

doi:10.19732/j.cnki.2096-6210.2021.06.005

您当前的位置：

首页 >

文章列表页 >

Comparative study on the utility of different radiomics feature selection methods in predicting survival of early-stage NSCLC patients

更新时间：2025-12-15

- Comparative study on the utility of different radiomics feature selection methods in predicting survival of early-stage NSCLC patients
- Vol. 30, Issue 6, Pages: 459-465(2021)
- 作者机构：
  
  1. 分子影像精准诊疗省部共建协同创新中心,山西,太原,030001
  2. 山西医科大学公共卫生学院,山西,太原,030001
  3. 山西医科大学第一医院核医学科,山西,太原,030001
- 作者简介：
- 基金信息：
- DOI：10.19732/j.cnki.2096-6210.2021.06.005
  CLC：
- Published Online：28 December 2021，
  
  Published：28 December 2021
- 稿件说明：
移动端阅览
王鑫超,崔曹哲,胡奕奕,等. 不同影像组学特征筛选方法对早期NSCLC患者生存预测效能的比较研究[J]. 肿瘤影像学, 2021, 30(6): 459-465 https://doi.

org/10.19732/j.cnki.2096-6210.2021.06.005
王鑫超,崔曹哲,胡奕奕,等. 不同影像组学特征筛选方法对早期NSCLC患者生存预测效能的比较研究[J]. 肿瘤影像学, 2021, 30(6): 459-465 https://doi. DOI： 10.19732/j.cnki.2096-6210.2021.06.005.

org/10.19732/j.cnki.2096-6210.2021.06.005 DOI：

摘要

目的：

探讨不同特征选择方法用于

F-脱氧葡萄糖（fluorodeoxyglucose，FDG）正电子发射体层成像（positron emission tomography，PET）/计算机体层成像（computed tomography，CT）影像组学特征在早期非小细胞肺癌（non-small-cell lung cancer，NSCLC）患者3年总生存期（overall survival，OS）预测中的效果。

方法：

回顾并分析81例2017年3月2018年9月在山西医科大学第一医院进行全身

F-FDG PET/CT扫描并至少随访3年81例的临床早期NSCLC［美国癌症联合委员会（American Joint Committee on Cancer，AJCC）分期Ⅰ、Ⅱ期］患者的临床及影像学资料。分别从PET及CT图像的感兴趣区（region of interest，ROI）中提取影像组学特征，利用最小绝对收缩和选择算子（least absolute shrinkage and selection operator，LASSO）算法、互信息（mutual information，MI）算法、递归特征消除（recursive feature elimination，RFE）算法以及Cox单因素分析法进行特征选择并构建Cox比例风险模型；通过内部验证后，使用一致性指数（C-index）评价不同模型预测NSCLC患者3年OS的能力。

结果：

影像组学特征与生存结局有较强相关性，可作为分析变量；相较其他特征选择方法，LASSO筛选的特征所构建Cox模型C-index最高（0.830.07）。

结论：

特征选择方法直接影响预测模型的预测能力，在利用影像组学特征构建模型时应寻找合适的特征选择方法以提高模型效能。

Abstract

Objective:

To investigate the utility of different feature selection methods for

F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomog

raphy (CT) radiomics features in predicting 3-year overall survival (OS) of patients with early-stage non-small cell lung cancer (NSCLC).

Methods:

From March 2017 to September 2018

the clinical and imaging data of 81 patients with clinically early-stage NSCLC

American Joint Committee on Cancer (AJCC) Ⅰ and Ⅱ stage

who underwent

F-FDG PET/CT imaging in The First Hospital of Shanxi Medical University and were followed up for at least 3 years were retrospectively analyzed. The radiomics features were extracted respectively from the region of interest (ROI) of PET and CT. The least absolute shrinkage and selection operator (LASSO) algorithm

mutual information (MI) algorithm

recursive feature elimination (RFE) algorithm and univariate Cox regression analysis are used to select features for constructing Cox proportional risk models. After internal validation

the consistency index (C-index) was used to evaluate the efficiencies of each model in predicting the 3-year OS in NSCLC patients.

Results:

Radiomic features were strongly correlated with survival outcomes and could be used as analysis variables; compared with other methods

the Cox proportional risk models constructed by LASSO has the highest C-index (0.830.07).

Conclusion:

The feature selection method directly affects the efficiencies of model

and a suitable method should be found to improve the efficiencies when using radiomic features.

关键词

Keywords

references

Views

1586

下载量

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

罗萨伊-多尔夫曼病 ¹⁸ F-FDG PET/CT检查与评估1例

Clinical value of ¹⁸ F-FDG PET/CT in ¹³¹ I treatment of metastases of differentiated thyroid cancer

Discussion on the optimal acquisition and reconstruction protocol of uMI780 PET/CT by using ¹⁸F-FDG

Related Author

邢岩

乔文礼

赵晋华

汪太松

LU Hankui

WANG Yang

SHENG Shiwei

WANG Lili

Related Institution

Department of Nuclear Medicine, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine

Department of Medical Equipment, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine

Shanghai United Imaging Healthcare Co.,Ltd.

AI问答

⁰