文献速递腹部介入CT

基于增强CT机器学习的肝细胞癌经动脉化疗栓塞获影像完全缓解后早期复发预测与风险分层

Early recurrence prediction and risk stratification of hepatocellular carcinoma after transarterial chemoembolization achieving imaging complete response based on contrast-enhanced CT machine learning.
PubMed 文献 · Abdom Radiol (NY) · 2026-03-03

本研究利用两个中心122例经TACE达影像完全缓解的肝细胞癌患者的临床与增强CT参数,构建并验证6种机器学习模型预测1年内早期复发,其中XGBoost模型基于13个参数表现最佳,内部和外部验证AUC分别为0.913和0.812;基于ECV、完整包膜、FIB-4指数、肿瘤大小、血小板与中性粒细胞比值及延迟期肿瘤CT值构建的Cox回归列线图可将患者分为三个风险亚组。

要点

  • XGBoost模型基于13个参数预测TACE后1年早期复发,内部验证AUC 0.913,外部验证AUC 0.812。
  • Cox列线图纳入ECV、完整包膜、FIB-4、肿瘤大小、血小板与中性粒细胞比值及延迟期肿瘤CT值。
  • 列线图在两个数据集中均实现显著风险分层(均p < 0.01),可将患者分为三个亚组。

核心结论

  • 基于XGBoost的早期复发预测模型可识别TACE获影像完全缓解后1年复发,Cox回归列线图可实现风险分层并将患者分为三个亚组。

研究设计与样本

  • 研究纳入两个中心122例接受TACE并达影像完全缓解的肝细胞癌患者,分为开发集(n = 100)和外部验证集(n = 22)。

诊断标准与关键数值

  • 以1年为界将患者分为早期复发组和非早期复发组。
  • 收集并筛选40个临床和增强CT参数。
  • 构建并比较6种机器学习模型,采用AUC和决策曲线分析(DCA)评估。
  • XGBoost模型基于13个参数表现最佳,内部验证AUC 0.913,外部验证AUC 0.812。
  • Cox回归列线图纳入ECV、完整包膜、FIB-4指数、肿瘤大小、血小板与中性粒细胞比值及延迟期肿瘤CT值,log-rank检验在两个数据集中均p < 0.01。

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

OBJECTIVES: To develop and validate machine learning (ML) models using clinical and contrast-enhanced CT (CECT) parameters to assess recurrence risk in hepatocellular carcinoma (HCC) after transarterial chemoembolization (TACE) achieving imaging complete response (CR).

METHODS: 122 HCC patients who underwent TACE and achieved imaging CR from two centers were divided into the development (n = 100) and external validation dataset (n = 22). Recurrence free survival (RFS) was tracked, and patients were categorized into early recurrence (ER) and non-ER groups based on a 1-year cutoff. Forty clinical and CECT parameters were collected and screened. Six ML models were constructed and compared using the area under the curve (AUC) and decision curve analysis (DCA). Key parameters were used to construct a Cox regression nomogram and stratify recurrence risk using log-rank test.

RESULTS: The extreme gradient boosting (XGBoost) model demonstrated the best predictive performance based on 13 parameters, with AUCs of 0.913 and 0.812 for the internal and external validation datasets. SHapley Additive exPlanations (SHAP) analysis identified the top 10 parameters. The Cox regression nomogram was constructed with ECV, complete capsule, FIB-4 index, tumor size, platelet-to-neutrophil ratio, and delayed phase tumor CT value. Log-rank test demonstrated significant risk stratification in both datasets (both p < 0.01).

CONCLUSION: The XGBoost-based ER prediction model identifies 1-year recurrence following TACE with imaging CR. The Cox regression nomogram enables risk stratification, dividing patients into three subgroups.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Liu L, Liu Y, Lin D, et al. Early recurrence prediction and risk stratification of hepatocellular carcinoma after transarterial chemoembolization achieving imaging complete response based on contrast-enhanced CT machine learning. Abdom Radiol (NY). 2026. PMID: 41774134. doi:10.1007/s00261-026-05436-x

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。