文献速递眼底病与葡萄膜炎

多中心前瞻性研究:机器学习模型预测息肉状脉络膜血管病变复发

Development and validation of a machine learning model to predict recurrence in polypoidal choroidal vasculopathy: a multicentre prospective study.
PubMed 文献 · Br J Ophthalmol · 2026-05-05

该多中心前瞻性研究基于OCT/OCTA生物标志物开发并验证预测息肉状脉络膜血管病变(PCV)复发的机器学习模型,纳入204眼,外部验证AUC 0.801–0.861,XGBoost模型表现最佳。SHAP分析显示息肉高度百分比变化、分支新生血管网面积变化和中心凹下脉络膜厚度变化为前三位预测因子。

要点

  • XGBoost模型在外部验证中AUC最高(0.861),可用于PCV复发预测。
  • 息肉高度百分比变化是SHAP排序首位的重要预测因子。
  • 模型基于OCT/OCTA影像生物标志物,辅助个体化治疗决策。

核心结论

  • XGBoost模型在外部验证集中展现出最佳预测性能(AUC 0.861),为PCV复发预测提供了可靠工具。

研究设计与样本

  • 多中心前瞻性研究,2019年6月至2023年12月在中国14家医院开展。
  • 纳入抗VEGF治疗后达到缓解的PCV患者,随访至少1年并系列OCT/OCTA成像。
  • 共204眼,其中125眼来自北京协和医院作为训练集,79眼来自中国其他13个中心作为外部验证集。
  • 采用LASSO回归(10折交叉验证)筛选预测特征,开发并评估五种分类器。

主要发现

  • 外部验证集中AUC范围为0.801至0.861,XGBoost模型AUC最高(0.861)。
  • SHAP分析显示息肉高度百分比变化、分支新生血管网面积变化和中心凹下脉络膜厚度变化为前三位显著预测因子。
  • 共筛选出10个特征用于模型开发。

关键数值与推荐

  • 五种分类器(逻辑回归、支持向量机、随机森林、K近邻和XGBoost)采用AUC、准确率、敏感性和特异性评估。
  • 外部验证集AUC范围0.801–0.861,XGBoost模型AUC为0.861。

临床怎么用

  • XGBoost模型可用于PCV患者抗VEGF治疗缓解后的复发预测,辅助个体化治疗决策和优化临床资源。

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)。

原文

AIM: This study aims to develop and validate machine learning models for predicting recurrence in polypoidal choroidal vasculopathy (PCV) patients using optical coherence tomography (OCT) and OCT angiography (OCTA) biomarkers. METHODS: This multicentre prospective study was conducted at 14 hospitals between June 2019 and December 2023. Patients who achieved remission after anti-vascular endothelial growth factor treatment were followed up for at least 1 year with serial OCT/OCTA imaging. Predictive features were selected using the least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation. Five classifiers (logistic regression, support vector machine, random forest, k-nearest neighbours and extreme gradient boosting (XGBoost)) were developed and evaluated using area under the curve (AUC), accuracy, sensitivity and specificity. Shapley additive explanations (SHAP) were applied for model interpretation and feature ranking. RESULTS: A total of 204 eyes were included, with 125 from Peking Union Medical College Hospital (training set) and 79 from 13 other centres across China (external validation set). Ten features were selected for model development. In the external validation set, AUCs ranged from 0.801 to 0.861, with the XGBoost model achieving the highest AUC (0.861). SHAP analysis revealed that the percent change in polyp height, the change in branching neovascular network area and the change in subfoveal choroidal thickness were the top three significant predictors. CONCLUSION: The XGBoost model demonstrated the best predictive performance, providing a reliable tool for recurrence prediction, aiding personalised treatment decisions and optimising clinical resources.

来源:Br J Ophthalmol,查看原文。版权归原作者所有。

引用

  1. Cheng S, Han R, Zhang W, et al. Development and validation of a machine learning model to predict recurrence in polypoidal choroidal vasculopathy: a multicentre prospective study. Br J Ophthalmol. 2026. PMID: 42086315. doi:10.1136/bjo-2025-328228

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。