文献速递胸部CT

用于鉴别≤3 cm微浸润性腺癌与浸润性腺癌的影像-影像组学模型:一项双中心回顾性研究

A Radiological-Radiomics model for differentiation between minimally invasive adenocarcinoma and invasive adenocarcinoma less than or equal to 3 cm: A two-center retrospective study.
PubMed 文献 · Eur J Radiol · 2024-05-27

该双中心回顾性研究纳入509例患者(522个病灶)的肺腺癌临床、病理及影像资料,中心1的392例(402个病灶)用于五折交叉验证训练与验证,中心2的117例(120个病灶)作为独立外部测试集。采用LASSO筛选特征、逻辑回归构建影像模型、影像组学模型及影像-影像组学联合模型预测浸润性腺癌。联合模型在训练集、验证集及外部测试集的AUC分别为0.918、0.906及0.894,作者认为其鉴别微浸润性腺癌与浸润性腺癌诊断性能优良。

要点

  • 联合模型在外部测试集鉴别微浸润性腺癌与浸润性腺癌的AUC为0.894(95%CI 0.824-0.964)。
  • 外部测试集敏感度84.8%、特异度78.6%、准确率83.3%。
  • 验证集AUC为0.906(95%CI 0.842-0.970),敏感度79.9%、特异度88.1%。

核心结论

  • 影像-影像组学联合模型在鉴别微浸润性腺癌与浸润性腺癌方面表现出优良的诊断性能,可为临床诊断和手术治疗方案提供一定参考。

研究设计与样本

  • 双中心回顾性研究,共509例患者(522个病灶)经手术病理诊断的肺腺癌;中心1的392例(402个病灶)采用五折交叉验证进行训练与验证,中心2的117例(120个病灶)作为独立外部测试集;采用LASSO筛选特征,逻辑回归构建影像模型、影像组学模型及影像-影像组学联合模型。

诊断标准与关键数值

  • 联合模型训练集预测浸润性腺癌:AUC 0.918(95%CI 0.889-0.947),敏感度80.3%,特异度88.2%,准确率82.1%。
  • 联合模型验证集预测浸润性腺癌:AUC 0.906(95%CI 0.842-0.970),敏感度79.9%,特异度88.1%,准确率81.8%。
  • 联合模型外部测试集预测浸润性腺癌:AUC 0.894(95%CI 0.824-0.964),敏感度84.8%,特异度78.6%,准确率83.3%。

临床应用与阅片要点

  • 作者认为联合模型可为临床诊断和手术治疗方案提供一定参考。

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: To develop a Radiological-Radiomics (R-R) combined model for differentiation between minimal invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IA) of lung adenocarcinoma (LUAD) and evaluate its predictive performance.

METHODS: The clinical, pathological, and imaging data of a total of 509 patients (522 lesions) with LUAD diagnosed by surgical pathology from 2 medical centres were retrospectively collected, with 392 patients (402 lesions) from center 1 trained and validated using a five-fold cross-validation method, and 117 patients (120 lesions) from center 2 serving as an independent external test set. The least absolute shrinkage and selection operator (LASSO) method was utilized to filter features. Logistic regression was used to construct three models for predicting IA, namely, Radiological model, Radiomics model, and R-R model. Also, receiver operating curve curves (ROCs) were plotted, generating corresponding area under the curve (AUC), sensitivity, specificity, and accuracy.

RESULTS: The R-R model for IA prediction achieved an AUC of 0.918 (95 % CI: 0.889-0.947), a sensitivity of 80.3 %, a specificity of 88.2 %, and an accuracy of 82.1 % in the training set. In the validation set, this model exhibited an AUC of 0.906 (95 % CI: 0.842-0.970), a sensitivity of 79.9 %, a specificity of 88.1 %, and an accuracy of 81.8 %. In the external test set, the AUC was 0.894 (95 % CI: 0.824-0.964), a sensitivity of 84.8 %, a specificity of 78.6 %, and an accuracy of 83.3 %.

CONCLUSION: The R-R model showed excellent diagnostic performance in differentiating MIA and IA, which can provide a certain reference for clinical diagnosis and surgical treatment plans.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Dong H, Xi Y, Liu K, et al. A Radiological-Radiomics model for differentiation between minimally invasive adenocarcinoma and invasive adenocarcinoma less than or equal to 3 cm: A two-center retrospective study. Eur J Radiol. 2024. PMID: 38820952. doi:10.1016/j.ejrad.2024.111532

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。