文献速递妇产影像泌尿生殖MR

基于常规MRI特征与影像组学的可解释机器学习模型鉴别子宫肉瘤与非典型子宫肌瘤

Interpretable Machine Learning Model for Differentiating Uterine Sarcoma From Atypical Leiomyoma Based on Conventional MRI Features and Radiomics.
PubMed 文献 · Acad Radiol · 2025-11-12

本研究回顾性纳入160例患者(子宫肉瘤47例、非典型子宫肌瘤113例),随机分为训练集112例和测试集48例,由两名盲法放射科医师评估10项盆腔MRI特征并提取T2WI与DWI影像组学特征,构建5种机器学习模型。多因素分析确定T2WI不均匀高信号、边界不清、子宫内膜腔中断和低ADC值为独立鉴别因素,XGBoost模型在训练集和测试集的AUC分别为0.991和0.909,SHAP分析显示ADC值贡献最大。

要点

  • ADC值是鉴别子宫肉瘤与非典型子宫肌瘤最具影响力的预测因子,其次为肿瘤边界、T2WI信号、影像组学评分和子宫内膜腔。
  • 多因素分析显示T2WI不均匀高信号(OR=43.767,P=0.021)、边界不清(OR=4.887,P=0.038)、子宫内膜腔中断(OR=15.947,P=0.003)和低ADC值(OR=0.026,P=0.009)为独立鉴别因素。
  • XGBoost模型表现最优,训练集AUC为0.991(95%CI:0.978-1.000),测试集AUC为0.909(95%CI:0.822-0.995)。

核心结论

  • 整合MRI生物标志物与影像组学的可解释机器学习模型为术前鉴别子宫肉瘤与非典型子宫肌瘤提供了透明且临床可操作的工具,通过SHAP量化特征贡献弥合机器学习的「黑箱」差距,作者认为有助于临床医师制定精准干预如合理手术规划以避免疑似子宫肉瘤的粉碎术。

研究设计与样本

  • 回顾性研究,纳入160例患者(子宫肉瘤47例、非典型子宫肌瘤113例),随机分为训练集112例和测试集48例,由两名盲法放射科医师评估10项盆腔MRI特征,从轴位T2WI和DWI序列提取影像组学特征并以LASSO回归筛选4个特征计算影像组学评分,训练逻辑回归、随机森林、XGBoost、支持向量机和高斯朴素贝叶斯5种机器学习模型,通过AUC、校准曲线和决策曲线分析评估性能,并以SHAP框架提供可解释可视化。

关键影像表现/征象

  • 多因素分析确定4项MRI鉴别因素:T2WI不均匀高信号(OR=43.767,P=0.021)、肿瘤边界不清(OR=4.887,P=0.038)、子宫内膜腔中断(OR=15.947,P=0.003)和低ADC值(OR=0.026,P=0.009)。
  • SHAP分析显示ADC值为最具影响力的预测因子,其次为肿瘤边界、T2WI信号强度、影像组学评分和子宫内膜腔。

诊断标准与关键数值

  • XGBoost模型在训练集AUC为0.991(95%CI:0.978-1.000),在测试集AUC为0.909(95%CI:0.822-0.995)。
  • 校准曲线显示预测结果与观察结果高度一致,决策曲线分析证实模型在概率阈值范围内具有临床效用。

临床应用与阅片要点

  • 作者认为该可解释机器学习框架通过SHAP值量化特征贡献,可增强临床医师信任并辅助制定精准干预如合理手术规划以避免疑似子宫肉瘤的粉碎术。
  • 阅片时可关注T2WI不均匀高信号、肿瘤边界不清、子宫内膜腔中断和低ADC值等鉴别特征。

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: This study aims to develop interpretable machine learning (ML) models by integrating conventional magnetic resonance imaging (MRI) features and radiomics to preoperatively differentiate uterine sarcoma (US) and atypical leiomyoma (ALM).

MATERIALS AND METHODS: In this retrospective study, 160 patients (47 US, 113 ALM) were randomized into training (n=112) and test (n=48) cohorts. Two blinded radiologists assessed 10 MRI features from pelvic MRI examinations, including tumor border morphology, T2-weighted image (T2WI) signal heterogeneity, uterine endometrial cavity, apparent diffusion coefficient (ADC) value, and other features. Significant MRI features were identified through univariable and multivariable logistic regression analyses. Radiomics features were extracted from axial T2WI and diffusion-weighted imaging (DWI) sequences, with least absolute shrinkage and selection operator regression identifying four discriminative features for radiomic score (radscore) calculation. Five ML models are as follows: logistic regression (LR), random forest (RF), eXtreme gradient boosting (XGBoost), support vector machine (SVM), and Gaussian Naive Bayes (GNB) were trained using significant MRI predictors and radscore. Model performance was evaluated via area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). The SHapley Additive exPlanation (SHAP) framework provided interpretable visualizations of feature contributions.

RESULTS: Multivariable analysis identified four MRI discriminators as follows: heterogeneous hyperintensity on T2WI (odds ratio [OR]=43.767, P=0.021), ill-defined tumor border (OR=4.887, P=0.038), interrupted uterine cavity (OR=15.947, P=0.003), and low ADC values (OR=0.026, P=0.009). The XGBoost model achieved superior performance, with AUCs of 0.991 (95% confidence interval [CI]: 0.978-1.000) and 0.909 (95% CI: 0.822-0.995) in training and test cohorts, respectively. SHAP analysis highlighted ADC value as the most influential predictor, followed by tumor border, signal intensity on T2WI, radscore, and uterine endometrial cavity. DCA confirmed clinical utility across probability thresholds, and calibration curves demonstrated strong agreement between predicted and observed outcomes.

CONCLUSION: Interpretable ML models integrating MRI biomarkers and radiomics provide a transparent and clinically actionable tool for preoperative differentiation of US and ALM. By quantifying feature contributions through SHAP and providing a transparent SHAP value, this framework bridges the "black-box" gap in ML, fostering clinicians trust and empowering clinicians to formulate precise interventions, such as appropriate surgical planning to avoid the morcellation of suspected US.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Yang Z, Sun W, Jiang Y, et al. Interpretable Machine Learning Model for Differentiating Uterine Sarcoma From Atypical Leiomyoma Based on Conventional MRI Features and Radiomics. Acad Radiol. 2025. PMID: 41233272. doi:10.1016/j.acra.2025.10.013

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。