文献速递神经技术与安全CT

人工智能在非增强CT上检测缺血性卒中:系统综述与Meta分析

Artificial Intelligence for Ischemic Stroke Detection in Non-contrast CT: A Systematic Review and Meta-analysis.
PubMed 文献 · Acad Radiol · 2025-09-25

该系统综述与Meta分析纳入38项研究(32项研究提取74个试验),评估AI模型在非增强CT上检测缺血性卒中的诊断准确性,并与临床医生比较。AI内部验证汇总敏感度91.2%、特异度96.0%,外部验证敏感度降至59.8%、特异度97.3%;临床医生内部验证敏感度44.1%、特异度85.5%,外部验证敏感度46.1%、特异度83.6%;临床医生借助AI后敏感度83.7%、特异度86.7%。作者结论认为AI内部验证表现可接受,但外部验证不足限制其泛化性和实际适用性。

要点

  • AI内部验证汇总敏感度91.2%、特异度96.0%,外部验证敏感度59.8%、特异度97.3%。
  • 临床医生内部验证敏感度44.1%、特异度85.5%,外部验证敏感度46.1%、特异度83.6%。
  • 临床医生借助AI后汇总敏感度83.7%、特异度86.7%。

核心结论

  • AI在非增强CT上检测缺血性卒中的内部验证表现可接受,但Meta分析观察到显著异质性,且外部验证不足限制其真实临床环境中的泛化性和实际适用性。

研究设计与样本

  • 系统综述与Meta分析,检索PubMed、Web of Science、Cochrane、IEEE Xplore和Embase截至2025年2月,纳入38项研究(32项研究提取74个试验),使用PROBAST评估偏倚风险,采用汇总敏感度、特异度和HSROC曲线进行Meta分析。

诊断标准与关键数值

  • AI内部验证汇总敏感度91.2%(95%CI: 87.6%-93.8%)、特异度96.0%(95%CI: 93.6%-97.6%)。
  • AI外部验证汇总敏感度59.8%(95%CI: 39.9%-76.9%)、特异度97.3%(95%CI: 93.2%-98.9%)。
  • 临床医生内部验证汇总敏感度44.1%(95%CI: 33.8%-55.0%)、特异度85.5%(95%CI: 68.4%-94.1%)。
  • 临床医生外部验证汇总敏感度46.1%(95%CI: 31.5%-61.3%)、特异度83.6%(95%CI: 62.8%-93.9%)。
  • 临床医生借助AI后汇总敏感度83.7%(95%CI: 53.0%-95.9%)、特异度86.7%(95%CI: 77.1%-92.6%)。

临床应用与阅片要点

  • 亚组分析显示,数据增强与模型敏感度较高相关(93.9%,95%CI: 90.2%-96.2%),迁移学习与模型敏感度较高相关(94.7%,95%CI: 92.0%-96.6%)。
  • 38项研究中22项(58%)被判定为高偏倚风险。
  • 敏感性分析和亚组分析识别出数据异质性的多个来源,包括偏倚风险和AI模型类型。

局限性

  • Meta分析观察到显著异质性。
  • 外部验证不足限制AI在真实临床环境中的泛化性和实际适用性。
  • 38项研究中22项(58%)被判定为高偏倚风险。
  • 敏感性分析和亚组分析识别出数据异质性的多个来源,包括偏倚风险和AI模型类型。

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 OBJECTIVE: We aim to conduct a systematic review and meta-analysis to objectively assess the diagnostic accuracy of artificial intelligence (AI) models for detecting ischemic stroke (IS) in non-contrast CT (NCCT), and to compare the diagnostic performance between AI and clinicians.

MATERIALS AND METHODS: Until February 2025, systematic searches were conducted in PubMed, Web of Science, Cochrane, IEEE Xplore, and Embase for studies using AI based on NCCT images from human subjects for IS detection or classification. The risk of bias was evaluated using the prediction model study risk of bias assessment tool (PROBAST). For meta-analysis, the pooled sensitivities, specificities, and hierarchical summary receiver operating characteristic (HSROC) curves were used.

RESULTS: A total of 38 studies, with 74 trials extracted from 32 studies were included. For AI performance, the pooled sensitivity and specificity were 91.2% (95%CI: 87.6%-93.8%) and 96.0% (95%CI: 93.6%-97.6%) for internal validation and 59.8% (95%CI:39.9%-76.9%) and 97.3% (95%CI: 93.2%-98.9%) for external validation. For clinicians' performance, the pooled sensitivity and specificity were 44.1% (95%CI: 33.8%-55.0%) and 85.5% (95%CI: 68.4%-94.1%) for internal validation and 46.1% (95%CI: 31.5%-61.3%) and 83.6% (95%CI: 62.8%-93.9%) for external validation. The pooled sensitivity and specificity increased to 83.7% (95%CI: 53.0%-95.9%) and 86.7% (95%CI: 77.1%-92.6%) for clinicians with AI assistance. The subgroup analysis results indicated that higher model sensitivity was associated with the data augmentation (93.9%, 95%CI: 90.2%-96.2%) and transfer learning (94.7%, 95%CI: 92.0%-96.6%). There were 22 of 38 (58%) studies that were judged to have high risk of bias. Sensitive analysis and subgroup analysis identified multiple sources of heterogeneity in the data, including risk of bias and AI model types.

CONCLUSION: Our study reveals that AI has an acceptable performance in detecting IS in NCCT in internal validation, although significant heterogeneity was observed in the meta-analysis. However, the generalizability and practical applicability of AI in real-world clinical settings remain limited due to insufficient external validation.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Shen W, Peng J, Lu J, et al. Artificial Intelligence for Ischemic Stroke Detection in Non-contrast CT: A Systematic Review and Meta-analysis. Acad Radiol. 2025. PMID: 41006084. doi:10.1016/j.acra.2025.09.010

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。