文献速递神经CT

基于非增强CT扫描的深度学习算法用于蛛网膜下腔出血病因分类的开发

Development of a deep-learning algorithm for etiological classification of subarachnoid hemorrhage using non-contrast CT scans.
PubMed 文献 · Eur Radiol · 2025-05-17

本研究开发并验证了一种基于非增强CT的深度学习算法,用于区分动脉瘤性蛛网膜下腔出血(aSAH)与非动脉瘤性蛛网膜下腔出血(naSAH)。回顾性纳入618例SAH患者,训练与内部验证集533例(aSAH=305,naSAH=228),外部测试集85例(aSAH=55,naSAH=30)。模型在内部验证集AUC为0.948(95%CI: 0.929-0.967),外部测试集AUC为0.914(95%CI: 0.889-0.940),外部测试集平均准确率0.887,优于低年资放射科医师(平均准确率0.836)。

要点

  • 非增强CT深度学习模型可区分动脉瘤性与非动脉瘤性SAH,外部测试AUC 0.914(95%CI: 0.889-0.940)。
  • 模型外部测试平均敏感度0.891、特异度0.880、准确率0.887、MCC 0.761。
  • 外部测试中模型平均准确率0.887,优于低年资放射科医师的0.836。

核心结论

  • 基于非增强CT的ResNet变体深度学习模型能准确识别SAH病因,在内部验证集AUC 0.948(95%CI: 0.929-0.967)、外部测试集AUC 0.914(95%CI: 0.889-0.940),并提升低年资放射科医师的诊断表现。

研究设计与样本

  • 回顾性研究,共纳入618例诊断为SAH的患者,分为训练与内部验证队列533例(aSAH=305,naSAH=228)和外部测试队列85例(aSAH=55,naSAH=30)。

关键影像表现/征象

  • 采用U-Net++架构自动分割出血区域,并以ResNet为基础的深度学习模型对SAH病因进行分类。

诊断标准与关键数值

  • 内部验证队列:平均敏感度0.898、特异度0.877、准确率0.889、MCC 0.777、AUC 0.948(95%CI: 0.929-0.967)。
  • 外部测试队列:平均敏感度0.891、特异度0.880、准确率0.887、MCC 0.761、AUC 0.914(95%CI: 0.889-0.940)。
  • 外部测试队列中,模型平均准确率0.887、MCC 0.761,优于低年资放射科医师的平均准确率0.836、MCC 0.660。

临床应用与阅片要点

  • 作者指出,该模型的高诊断性能突显其在急诊环境中支持快速精准临床决策的潜力。
  • 作者指出,AI驱动的非增强CT分析为术前SAH诊断提供快速、经济、无创的解决方案,有助于早期识别需动脉瘤手术的患者并减少非动脉瘤病例不必要的血管造影。

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: This study aims to develop a deep learning algorithm for differentiating aneurysmal subarachnoid hemorrhage (aSAH) from non-aneurysmal subarachnoid hemorrhage (naSAH) using non-contrast computed tomography (NCCT) scans.

METHODS: This retrospective study included 618 patients diagnosed with SAH. The dataset was divided into a training and internal validation cohort (533 cases: aSAH = 305, naSAH = 228) and an external test cohort (85 cases: aSAH = 55, naSAH = 30). Hemorrhage regions were automatically segmented using a U-Net + + architecture. A ResNet-based deep learning model was trained to classify the etiology of SAH.

RESULTS: The model achieved robust performance in distinguishing aSAH from naSAH. In the internal validation cohort, it yielded an average sensitivity of 0.898, specificity of 0.877, accuracy of 0.889, Matthews correlation coefficient (MCC) of 0.777, and an area under the curve (AUC) of 0.948 (95% CI: 0.929-0.967). In the external test cohort, the model demonstrated an average sensitivity of 0.891, specificity of 0.880, accuracy of 0.887, MCC of 0.761, and AUC of 0.914 (95% CI: 0.889-0.940), outperforming junior radiologists (average accuracy: 0.836; MCC: 0.660).

CONCLUSION: The study presents a deep learning architecture capable of accurately identifying SAH etiology from NCCT scans. The model's high diagnostic performance highlights its potential to support rapid and precise clinical decision-making in emergency settings.

KEY POINTS: Question Differentiating aneurysmal from naSAH is crucial for timely treatment, yet existing imaging modalities are not universally accessible or convenient for rapid diagnosis. Findings A ResNet-variant-based deep learning model utilizing non-contrast CT scans demonstrated high accuracy in classifying SAH etiology and enhanced junior radiologists' diagnostic performance. Clinical relevance AI-driven analysis of non-contrast CT scans provides a fast, cost-effective, and non-invasive solution for preoperative SAH diagnosis. This approach facilitates early identification of patients needing aneurysm surgery while minimizing unnecessary angiography in non-aneurysmal cases, enhancing clinical workflow efficiency.

来源:Eur Radiol,查看原文。版权归原作者所有。

引用

  1. Chen L, Wang X, Li Y, et al. Development of a deep-learning algorithm for etiological classification of subarachnoid hemorrhage using non-contrast CT scans. Eur Radiol. 2025. PMID: 40382487. doi:10.1007/s00330-025-11666-2

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。