文献速递妇产影像MR

基于ADC图影像组学列线图术前鉴别I型与II型子宫内膜癌并辅助诊刮

Radiomics nomogram in aiding preoperatively dilatation and curettage in differentiating type II and type I endometrial cancer.
PubMed 文献 · Clin Radiol · 2022-10-01

回顾性纳入875例子宫内膜癌患者,按1:1随机分为训练组437例和测试组438例,从ADC图提取影像组学特征并结合临床独立危险因素构建列线图。训练组中列线图AUC为0.93,高于诊刮的0.85和临床因素的0.70;测试组分别为0.91、0.82和0.64。结论认为基于ADC的影像组学列线图可改善诊刮对I型与II型子宫内膜癌的术前鉴别。

要点

  • ADC图影像组学列线图术前鉴别I/II型子宫内膜癌的AUC在训练组和测试组分别为0.93和0.91。
  • 列线图鉴别效能优于诊刮病理(训练组AUC 0.85,测试组0.82)。
  • DCA、NRI和IDI均提示列线图具有临床可用性,可辅助诊刮术前分型。

核心结论

  • 基于ADC图的影像组学列线图可改善诊刮结果,用于术前鉴别II型与I型子宫内膜癌。

研究设计与样本

  • 回顾性纳入875例子宫内膜癌患者,按1:1随机分为训练组437例和测试组438例。
  • 从ADC图提取并筛选影像组学特征,采用多变量逻辑回归识别独立临床危险因素。
  • 将影像组学特征与独立临床危险因素整合构建ADC基于的影像组学列线图。
  • 采用DCA评估临床有用性,计算NRI和IDI比较列线图与诊刮的判别性能。

诊断标准与关键数值

  • 训练组AUC:临床危险因素0.70(95%CI 0.64-0.76)、诊刮0.85(95%CI 0.80-0.89)、ADC影像组学列线图0.93(95%CI 0.90-0.96)。
  • 测试组AUC:临床危险因素0.64(95%CI 0.57-0.71)、诊刮0.82(95%CI 0.77-0.87)、ADC影像组学列线图0.91(95%CI 0.87-0.95)。

临床应用与阅片要点

  • 术前可基于ADC图提取影像组学特征并联合临床危险因素,辅助诊刮对I型与II型子宫内膜癌进行分型。
  • 列线图判别性能优于单独诊刮结果,DCA、NRI和IDI支持其临床可用性。

局限性

  • 原文为回顾性研究设计,摘要未报告外部验证队列。

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

AIM: To established a radiomics nomogram for improving the dilatation and curettage (D&C) result in differentiating type II from type I endometrial cancer (EC) preoperatively.

MATERIAL AND METHODS: EC patients (n=875) were enrolled retrospectively and divided randomly into a training cohort (n=437) and a test cohort (n=438), according to the ratio of 1:1. Radiomics signatures were extracted and selected from apparent diffusion coefficient (ADC) maps. A multivariate logistic regression analysis was used to identify the independent clinical risk factors. An ADC based-radiomics nomogram was built by integrating the selected radiomics signatures and the independent clinical risk factors. Decision curve analysis (DCA) was conducted to determine the clinical usefulness of the radiomics nomogram. The net reclassification index (NRI) and total integrated discrimination index (IDI) were calculated to compare the discrimination performances between the radiomics nomogram and the D&C result.

RESULTS: Receiver operating characteristic (ROC) curves showed that the clinical risk factors, the D&C, and the ADC based-radiomics nomogram yielded areas under the ROC curves (AUCs) of 0.70 (95% CI: 0.64-0.76), 0.85 (95% CI: 0.80-0.89), and 0.93 (95% CI: 0.90-0.96) in the training cohort and 0.64 (95% CI: 0.57-0.71), 0.82 (95% CI: 0.77-0.87) and 0.91 (95% CI: 0.87-0.95) in the test cohort, respectively. The DCA, NRI, and IDI demonstrated the clinically usefulness of the ADC based-radiomics nomogram.

CONCLUSION: The ADC-based radiomics nomogram could be used to improve the D&C result in differentiating type II from type I EC preoperatively.

来源:Clin Radiol,查看原文。版权归原作者所有。

引用

  1. Liu, Yan, Li, et al. Radiomics nomogram in aiding preoperatively dilatation and curettage in differentiating type II and type I endometrial cancer. Clin Radiol. 2022. PMID: 36192204. doi:10.1016/j.crad.2022.08.139

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。