文献速递胸部CT

基于机器学习的原发性肺腺癌STAS术前风险预测:一种SHAP可解释模型

Preoperative Prediction of STAS Risk in Primary Lung Adenocarcinoma Using Machine Learning: An Interpretable Model with SHAP Analysis.
PubMed 文献 · Acad Radiol · 2025-03-25

这项多中心回顾性研究纳入1237例病理确诊的原发性肺腺癌患者,利用术前CT影像特征和临床病理数据构建7种机器学习模型预测STAS,其中XGB模型在训练集、内部测试集和外部验证集的AUC分别为0.973、0.862和0.842/0.810,SHAP分析识别结节类型、癌胚抗原、最大结节直径和分叶征为关键预测特征。

要点

  • XGB模型预测STAS的AUC在训练集为0.973,内部测试集为0.862,外部验证集为0.842/0.810。
  • SHAP分析显示结节类型、癌胚抗原、最大结节直径和分叶征是预测STAS的关键特征。
  • 逻辑回归分析证实上述特征为STAS的独立危险因素。

核心结论

  • XGB模型对STAS具有较高的预测准确性和可解释性,整合常规可得的临床和影像特征,为术前风险分层提供了实用有效的工具,支持原发性肺腺癌的个体化手术规划。

研究设计与样本

  • 多中心回顾性研究,纳入三家医院1237例病理确诊的原发性肺腺癌患者;中心1的932例分为训练集652例和内部测试集280例,中心2的165例和中心3的140例构成外部验证集;使用Boruta和最小绝对收缩与选择算子回归筛选CT影像特征和临床变量,开发7种机器学习模型并进行五折交叉验证。

关键影像表现/征象

  • SHAP分析识别结节类型、癌胚抗原、最大结节直径和分叶征为预测STAS的关键特征。
  • 逻辑回归分析证实结节类型、癌胚抗原、最大结节直径和分叶征为独立危险因素。

诊断标准与关键数值

  • XGB模型预测STAS的AUC在训练集为0.973、内部测试集为0.862、外部验证集为0.842/0.810。
  • 模型性能评估指标包括F1分数、召回率、精确率、特异度、敏感度和受试者工作特征曲线下面积(AUC)。

临床应用与阅片要点

  • 模型整合常规可得的临床和影像特征,可用于原发性肺腺癌术前STAS风险分层,支持个体化手术规划。
  • 阅片时可关注结节类型、最大结节直径和分叶征等影像特征,结合癌胚抗原水平辅助评估STAS风险。

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

BACKGROUND: Accurate preoperative prediction of spread through air spaces (STAS) in primary lung adenocarcinoma (LUAD) is critical for optimizing surgical strategies and improving patient outcomes.

OBJECTIVE: To develop a machine learning (ML) based model to predict STAS using preoperative CT imaging features and clinicopathological data, while enhancing interpretability through shapley additive explanations (SHAP) analysis.

MATERIALS AND METHODS: This multicenter retrospective study included 1237 patients with pathologically confirmed primary LUAD from three hospitals. Patients from Center 1 (n=932) were divided into a training set (n=652) and an internal test set (n=280). Patients from Centers 2 (n=165) and 3 (n=140) formed external validation sets. CT imaging features and clinical variables were selected using Boruta and least absolute shrinkage and selection operator regression. Seven ML models were developed and evaluated using five-fold cross-validation. Performance was assessed using F1 score, recall, precision, specificity, sensitivity, and area under the receiver operating characteristic curve (AUC).

RESULTS: The Extreme Gradient Boosting (XGB) model achieved AUCs of 0.973 (training set), 0.862 (internal test set), and 0.842/0.810 (external validation sets). SHAP analysis identified nodule type, carcinoembryonic antigen, maximum nodule diameter, and lobulated sign as key features for predicting STAS. Logistic regression analysis confirmed these as independent risk factors.

CONCLUSION: The XGB model demonstrated high predictive accuracy and interpretability for STAS. By integrating widely available clinical and imaging features, this model offers a practical and effective tool for preoperative risk stratification, supporting personalized surgical planning in primary LUAD management.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Wang P, Cui J, Du H, et al. Preoperative Prediction of STAS Risk in Primary Lung Adenocarcinoma Using Machine Learning: An Interpretable Model with SHAP Analysis. Acad Radiol. 2025. PMID: 40140276. doi:10.1016/j.acra.2025.03.005

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。