基于CT的影像组学在儿童细菌性肺炎和肺炎所致胸腔积液鉴别诊断中的价值
The value of CT-based radiomics for differentiation of pleural effusions in bacterial pneumonia and pneumonia in children.
作者信息
Li Jun, Si Jiajun, Yang Yanlin, Zhang Li, Deng Yushan, Ding Hao, Chen Xin, He Ling
机构信息
Department of Radiology, Children's Hospital of Chongqing Medical University, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Engineering Research Center of Stem Cell Therapy, Chongqing, China.
出版信息
Transl Pediatr. 2025 Jan 24;14(1):70-79. doi: 10.21037/tp-24-364. Epub 2025 Jan 21.
BACKGROUND
Bacterial pathogens and are the two main pathogens that cause community-acquired pneumonia complicated with pleural effusion (PE) in children, it is important to accurately differentiate between these two types of effusions. The aim of this study was to explore the feasibility and value of a radiomics approach based on non-contrast chest computed tomography (CT) scans in the differentiation of bacterial pneumonia PE (BPPE) and parapneumonic effusion (MPPE) in children.
METHODS
The clinical and CT imaging data of hospitalized children with PE detected by chest CT scans from December 2020 to December 2023 were retrospectively collected. A total of 167 cases of BPPE and 368 cases of MPPE were included, and all cases were randomly divided into a training set and a test set in the ratio of 7:3. The region of interest (ROI) was manually segmented in images of non-contrast chest CT scans to extract radiomics features. The optimal radiomics features were screened using Select K Best, max-relevance and min-redundancy (mRMR), least absolute shrinkage and selection operator (LASSO). Logistic regression (LR) was selected to construct the radiomics model. The receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC), 95% confidence interval (CI), sensitivity, specificity, and accuracy were calculated to evaluate the model performance.
RESULTS
A total of 2,264 radiomics features were extracted from each ROI, seven optimal features were finally selected. The AUC in the training set was 0.942 (95% CI: 0.917-0.967), with sensitivity, specificity, accuracy and precision of 89.9%, 82.1%, 87.4% and 91.7%, respectively. The AUC in the test set was 0.917 (95% CI: 0.868-0.965), with sensitivity, specificity, accuracy and precision of 87.4%, 80.0%, 85.1% and 90.7%, respectively.
CONCLUSIONS
The model based on CT radiomics demonstrates the potential to identify BPPE and MPPE in children and provides a new direction for future research.
背景
细菌性病原体和[此处原文缺失部分内容]是导致儿童社区获得性肺炎合并胸腔积液（PE）的两种主要病原体，准确区分这两种类型的胸腔积液非常重要。本研究的目的是探讨基于非增强胸部计算机断层扫描（CT）的放射组学方法在鉴别儿童细菌性肺炎胸腔积液（BPPE）和类肺炎性胸腔积液（MPPE）中的可行性和价值。
方法
回顾性收集2020年12月至2023年12月期间因胸部CT扫描检测出患有PE的住院儿童的临床和CT影像数据。共纳入167例BPPE和368例MPPE病例，所有病例按7:3的比例随机分为训练集和测试集。在非增强胸部CT扫描图像中手动分割感兴趣区域（ROI）以提取放射组学特征。使用最佳子集选择（Select K Best）、最大相关最小冗余（mRMR）、最小绝对收缩和选择算子（LASSO）筛选最佳放射组学特征。选择逻辑回归（LR）构建放射组学模型。绘制受试者工作特征（ROC）曲线，并计算曲线下面积（AUC）、95%置信区间（CI）、敏感性、特异性和准确性以评估模型性能。
结果
从每个ROI中总共提取了2264个放射组学特征，最终选择了7个最佳特征。训练集中的AUC为0.942（95%CI：0.917 - 0.967），敏感性、特异性、准确性和精确性分别为89.9%、82.1%、87.4%和91.7%。测试集中的AUC为0.917（95%CI：0.868 - 0.965），敏感性、特异性、准确性和精确性分别为87.4%、80.0%、85.1%和90.7%。
结论
基于CT放射组学的模型显示出鉴别儿童BPPE和MPPE的潜力，并为未来研究提供了新方向。
Zotero 插件