开发预测模型以识别多发性囊状动脉瘤患者中导致蛛网膜下腔出血的责任颅内动脉瘤
本研究纳入四个中心2018年5月至2023年12月的207例SAH患者共460个颅内动脉瘤,随机分为训练集(80%)和内部验证集(20%),并另用其他四个中心65例患者共147个动脉瘤作为外部验证集,开发整合动脉瘤形态特征的机器学习模型来识别导致SAH的责任动脉瘤,并与依赖出血模式和最大动脉瘤大小的传统预测标志物比较。13个机器学习模型中表现最佳的高斯过程、逻辑回归和二次判别分析模型在训练集、内部验证集和外部验证集的AUC分别为0.912、0.894、0.890,0.869、0.872、0.853和0.898、0.892、0.897,DeLong检验显示这些模型间无显著差异,但均优于传统预测标志物(P < 0.001)。
要点
- 整合动脉瘤形态特征的机器学习模型可较准确识别多发性颅内动脉瘤中的责任动脉瘤。
- 传统出血模式和最大动脉瘤大小预测责任动脉瘤的AUC仅0.488-0.523,接近随机水平。
- 最佳机器学习模型在外部验证集AUC为0.892-0.898,均优于传统预测标志物(P < 0.001)。
核心结论
- 整合多个形态特征的机器学习模型能在多发性颅内动脉瘤患者中准确预测导致SAH的责任动脉瘤,且优于传统预测标志物,从而有助于及时有效的治疗。
研究设计与样本
- 研究论文:纳入四个医院2018年5月至2023年12月的207例SAH患者共460个颅内动脉瘤,随机分为训练集(80%)和内部验证集(20%),另用其他四个医院65例患者共147个动脉瘤作为外部验证集,基于CTA开发13个机器学习模型并与依赖出血模式和最大动脉瘤大小的传统预测标志物比较。
诊断标准与关键数值
- 出血模式和最大动脉瘤大小在训练集AUC为0.496-0.505、内部验证集0.502-0.523、外部验证集0.488-0.498;高斯过程模型训练集AUC 0.912(95%CI 0.881-0.943)、内部验证集0.869(95%CI 0.798-0.941)、外部验证集0.898(95%CI 0.848-0.947)。
临床应用与阅片要点
- 作者认为机器学习模型可辅助识别多发性动脉瘤中的责任动脉瘤,且所有模型均优于传统预测标志物(P < 0.001)。
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
PURPOSE: To develop and test machine learning (ML) models using computed tomography angiography to identify the intracranial aneurysm (IA) responsible for subarachnoid hemorrhage (SAH) accurately in patients with multiple saccular IAs and to determine whether these models outperform traditional predictive markers.
MATERIALS AND METHODS: Two hundred seven SAH patients with 460 IAs from four hospitals were included from May 2018-December 2023 and randomly divided into training (80%) and internal validation (20%) sets. Additionally, an external validation set comprising 65 patients with 147 IAs from other four hospitals was used. The predictive models were developed using ML methods that integrated the morphological features of IAs (e.g., size and shape) to identify the responsible IA. These models were then compared with traditional predictive markers that relies on hemorrhage patterns and the maximum IA size.
RESULTS: The areas under the curves (AUCs) for the hemorrhage patterns and the maximum IA size were 0.496-0.505, 0.502-0.523, and 0.488-0.498 in the training, internal validation, and external validation sets, respectively. Among the 13 ML models, the best-performing models were the Gaussian process, logistic regression, and quadratic discriminant analysis models, with AUCs of 0.912 [95 % confidence interval (CI): 0.881-0.943], 0.894 (95 % CI: 0.861-0.928), and 0.890 (95 % CI: 0.756-0.924), respectively, for the training set; 0.869 (95 % CI: 0.798-0.941), 0.872 (95 % CI: 0.802-0.942), and 0.853 (95 % CI: 0.778-0.929), respectively, for the internal validation set; and 0.898 (95 % CI: 0.848-0.947), 0.892 (95 % CI: 0.840-0.943), and 0.897 (95 % CI: 0.847-0.947), respectively, for the external validation set. DeLong tests revealed no significant differences among these models, but all the models outperformed traditional predictive markers (P < 0.001).
CONCLUSION: ML models that integrate multiple morphological features can predict the IA responsible for SAH accurately in patients with multiple IAs. These models outperform traditional predictive markers in identifying the responsible IA, thereby facilitating prompt and effective treatment.
引用
- Wang H, Kong JF, Wen L, et al. Development of predictive models to identify the intracranial aneurysm responsible for subarachnoid hemorrhage in patients with multiple saccular aneurysms. Eur J Radiol. 2025. PMID: 41101005. doi:10.1016/j.ejrad.2025.112466
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。