开发高性能超声预测模型诊断子宫内膜癌:利用SHAP分析的可解释XGBoost算法
本研究基于2021至2023年病理证实的经阴道或经直肠超声患者,回顾性开发并验证XGBoost机器学习模型预测子宫内膜及宫腔恶性病变。测试集中绝经前组AUC为0.845,敏感度0.588、特异度0.923;绝经后组AUC为0.968,敏感度0.895、特异度0.931。SHAP分析识别出子宫内膜-肌层交界、内膜厚度、回声、彩色多普勒血流评分及血管模式为关键预测因子。
要点
- 绝经后组模型AUC达0.968,敏感度0.895、特异度0.931,性能优于绝经前组
- 绝经前组敏感度仅0.588,特异度0.923,需注意漏诊可能
- SHAP分析提示子宫内膜-肌层交界和内膜厚度为最重要预测因子
核心结论
- 基于XGBoost的模型预测性能优异,尤其在绝经后患者中;SHAP分析通过识别关键超声预测因子增强了可解释性。
研究设计与样本
- 回顾性研究,纳入2021至2023年经阴道或经直肠超声且病理证实的患者1080例,其中6例子宫内膜不可测量,剩余1074例中绝经前641例、绝经后433例;数据集按7:3随机分为训练集和验证集;采用LASSO回归进行特征选择并开发XGBoost模型;超声特征按IETA术语描述。
关键影像表现/征象
- 超声特征采用国际子宫内膜肿瘤分析(IETA)术语进行表征。
- SHAP分析识别绝经前女性关键预测因子:子宫内膜-肌层交界、子宫内膜厚度、子宫内膜回声、彩色多普勒血流评分和血管模式。
- SHAP分析识别绝经后女性关键预测因子:子宫内膜厚度、子宫内膜-肌层交界、子宫内膜回声和彩色多普勒血流评分。
诊断标准与关键数值
- 测试集绝经前组:AUC 0.845(95%CI 0.781-0.909),敏感度0.588(95%CI 0.442-0.722),特异度0.923(95%CI 0.863-0.959)。
- 测试集绝经后组:AUC 0.968(95%CI 0.944-0.992),敏感度0.895(95%CI 0.778-0.956),特异度0.931(95%CI 0.839-0.974)。
- 模型性能通过ROC分析、校准、决策曲线分析、敏感度、特异度和准确率评估。
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
OBJECTIVES: To develop and validate an ultrasonography-based machine learning (ML) model for predicting malignant endometrial and cavitary lesions.
METHODS: This retrospective study was conducted on patients with pathologically confirmed results following transvaginal or transrectal ultrasound from 2021 to 2023. Endometrial ultrasound features were characterized using the International Endometrial Tumor Analysis (IETA) terminology. The dataset was ranomly divided (7:3) into training and validation sets. LASSO (least absolute shrinkage and selection operator) regression was applied for feature selection, and an extreme gradient boosting (XGBoost) model was developed. Performance was assessed via receiver operating characteristic (ROC) analysis, calibration, decision curve analysis, sensitivity, specificity, and accuracy.
RESULTS: Among 1080 patients, 6 had a non-measurable endometrium. Of the remaining 1074 cases, 641 were premenopausal and 433 postmenopausal. Performance of the XGBoost model on the test set: The area under the curve (AUC) for the premenopausal group was 0.845 (0.781-0.909), with a relatively low sensitivity (0.588, 0.442-0.722) and a relatively high specificity (0.923, 0.863-0.959); the AUC for the postmenopausal group was 0.968 (0.944-0.992), with both sensitivity (0.895, 0.778-0.956) and specificity (0.931, 0.839-0.974) being relatively high. SHapley Additive exPlanations (SHAP) analysis identified key predictors: endometrial-myometrial junction, endometrial thickness, endometrial echogenicity, color Doppler flow score, and vascular pattern in premenopausal women; endometrial thickness, endometrial-myometrial junction, endometrial echogenicity, and color Doppler flow score in postmenopausal women.
CONCLUSION: The XGBoost-based model exhibited excellent predictive performance, particularly in postmenopausal patients. SHAP analysis further enhances interpretability by identifying key ultrasonographic predictors of malignancy.
引用
- Lai H, Wu Q, Weng Z, et al. Development of a High-Performance Ultrasound Prediction Model for the Diagnosis of Endometrial Cancer: An Interpretable XGBoost Algorithm Utilizing SHAP Analysis. J Ultrasound Med. 2025. PMID: 41024607. doi:10.1002/jum.70082
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。