基于多机器学习与生境影像组学的可解释术后胰腺癌预后模型:多中心研究
本回顾性研究纳入455例可切除胰腺癌(RPC)患者,按2018年1月至2024年7月划分为训练集与外部测试集,利用生境影像组学将肿瘤分割为亚区并结合随机生存森林(RSF)等七种机器学习模型预测无复发生存(RFS)。基于生境影像组学的RSF模型在训练集C-index为0.828、两个外部测试集为0.702与0.680,优于全肿瘤影像组学(p<0.05),并具备可解释性。
要点
- 生境影像组学RSF模型训练集C-index 0.828,两个外部测试集分别为0.702与0.680,优于全肿瘤影像组学(p<0.05)。
- 第一个测试集0.5、1、2年时间依赖ROC的AUC分别为0.71、0.83、0.79;第二个测试集分别为0.65、0.79、0.75。
- 两个外部测试集中预测低风险组RFS均显著长于预测高风险组(均p<0.05)。
核心结论
- 基于生境影像组学的RSF机器学习模型提高了术后RPC的预后准确性与可解释性,为个体化管理提供了有前景的工具。
研究设计与样本
- 回顾性研究,共纳入455例可切除胰腺癌患者,按2018年1月至2024年7月划分为训练集与外部测试集。
- 采用生境影像组学将肿瘤分割为亚区以捕捉局部异质性,比较包括随机生存森林(RSF)在内的七种机器学习模型,以Harrell C-index评价。
- 最优模型进一步行时间依赖ROC与Kaplan-Meier(KM)分析,并应用SHAP与SurvLIME增强模型可解释性。
关键影像表现/征象
- 可解释性分析识别出的关键变量包括生境亚区1的Feature 1、Feature 5、Feature 2、Feature 4,以及生境亚区3的Feature 3。
诊断标准与关键数值
- 基于生境影像组学的RSF模型C-index:训练集0.828,第一个外部测试集0.702,第二个外部测试集0.680,优于全肿瘤影像组学(p<0.05)。
- 时间依赖ROC的AUC:第一个测试集0.5年0.71、1年0.83、2年0.79;第二个测试集0.5年0.65、1年0.79、2年0.75。
- KM分析显示两个外部测试集中预测低风险组RFS均显著长于预测高风险组(均p<0.05)。
临床应用与阅片要点
- 原文未据此给出推荐。
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
RATIONALE AND OBJECTIVES: Accurate risk stratification is critical for guiding personalized treatment in resectable pancreatic cancer (RPC). This retrospective study assessed the utility of habitat radiomics for predicting recurrence-free survival (RFS) in RPC patients.
MATERIALS AND METHODS: A total of 455 RPC patients were divided into training and external test sets from January 2018 to July 2024. Tumors were segmented into subregions using habitat radiomics to capture localized heterogeneity. Seven machine learning models, including random survival forest (RSF), were compared using Harrell's C-index. The optimal model underwent further validation through time-dependent ROC and Kaplan-Meier (KM) analyses. Shapley additive explanations (SHAP) and survival local interpretable model-agnostic explanations (SurvLIME) were applied to enhance model interpretability.
RESULTS: The RSF model based on habitat radiomics achieved a C-index of 0.828 in the training cohort and 0.702, 0.680 in external test sets, outperforming whole-tumor radiomics (p<0.05). Time-dependent ROC analysis showed AUCs of 0.71, 0.83, and 0.79 at 0.5, 1, and 2 years in the first test set, and 0.65, 0.79, and 0.75 in the second test set. KM analysis revealed that the predicted low-risk groups had significantly longer RFS compared to the predicted high-risk groups in both external test sets (all p<0.05). Interpretability analysis identified key variables, including Feature 1, Feature 5, Feature 2, and Feature 4 from Habitat Subregion 1, and Feature 3 from Habitat Subregion 3.
CONCLUSION: The habitat radiomics RSF machine learning model improves prognostic accuracy and interpretability for postoperative RPC, providing a promising tool for personalized management.
引用
- Gu Q, Xing Y, Hu X, et al. Interpretable Prognostic Modeling for Postoperative Pancreatic Cancer Using Multi-machine Learning and Habitat Radiomics: A Multi-center Study. Acad Radiol. 2025. PMID: 40328538. doi:10.1016/j.acra.2025.04.025
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。