文献速递胸部CT

人工智能作为肺癌筛查首读的潜力研究

Potential for AI as first reader in lung cancer screening.
PubMed 文献 · Eur J Radiol · 2025-11-19

该研究回顾性分析BioMILD试验4104例基线低剂量CT,按LungRADS v1.1将原始阅片分为阴性与阳性,由AI软件分配类别,以2年内肺癌诊断为参考标准。AI敏感性与人工相当但特异性较低,AI高阴性预测值可能支持其作为肺癌筛查首读。

要点

  • AI与人工对LDCT阴性与阳性判读的一致率为83.5%(Kw 0.47;95%CI 0.43-0.50)。
  • AI敏感性91.2%与人工89.7%无显著差异(P=0.5637),特异性AI 75.7%低于人工90.0%(P<0.0001)。
  • AI作为首读预计可减少74.7%的LDCT阅片工作量。

核心结论

  • AI阅片敏感性91.2%与人工89.7%相当(P=0.5637),但特异性75.7%低于人工90.0%(P<0.0001)。
  • AI高阴性预测值99.8%可能支持其作为肺癌筛查首读。

研究设计与样本

  • 回顾性分析BioMILD试验的4104例基线低剂量CT。
  • 由放射科医师按LungRADS v1.1将原始阅片回顾性分为阴性(类别1、2)和阳性(类别3、4),并由AI软件进行类别分配。
  • 以2年内肺癌诊断为参考标准评估敏感度、特异度、阴性预测值和阳性预测值。
  • 采用Fleiss-Cohen加权k-Cohen指数(Kw)及95%CI衡量阅片者间一致性。

诊断标准与关键数值

  • LungRADS v1.1类别1、2归为阴性,类别3、4归为阳性。
  • AI敏感度91.2%、特异度75.7%、阳性预测值6.0%、阴性预测值99.8%。
  • 人工阅片敏感度89.7%、特异度90.0%、阳性预测值13.1%、阴性预测值99.8%。
  • 敏感度比较P=0.5637,特异度比较P<0.0001,阳性预测值比较P<0.0001,阴性预测值比较P=0.9351。
  • AI与人工判读一致率83.5%(Kw 0.47;95%CI 0.43-0.50)。

临床应用与阅片要点

  • AI作为首读预计可减少74.7%的LDCT阅片工作量。
  • AI高阴性预测值99.8%可能支持其作为肺癌筛查首读,但特异性低于人工阅片。

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 retrospectively assess the agreement between human and automated AI-based readings for low-dose computed tomography (LDCT) outcomes according to LungRADS v1.1 in lung cancer screening (LCS); to test the diagnostic performance of both readings.

METHODS: We included 4104 baseline LDCTs from the BioMILD trial. Original readings were retrospectively classified into "negative" (LungRADSv1.1 categories 1, 2) and "positive" (categories 3, 4) by a radiologist and analyzed by AI software for category assignment. Diagnosis of lung cancer (LC) at 2 years served as reference standard to assess sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of both human and AI. Agreement between readers was measured by the k-Cohen Index with Fleiss-Cohen weights (Kw) with 95 % CI.

RESULTS: Median age of participants was 60 years; 60.8 % were male and 79.2 % current smokers; 68/4104 (1.7 %) were diagnosed with LC; 6/68 (8.8 %) and 7/68 (10.3 %) LDCT were classified as negative by AI and human reading, respectively. The agreement between human and AI readings for negative and positive LDCTs was 83.5 % (Kw 0.47; 95 %CI: 0.43-0.50). Sensitivity and specificity were 91.2 % and 75.7 % for AI, and 89.7 % and 90.0 % for human reading (p-value 0.5637 and < 0.0001). PPV and NPV were 6.0 % and 99.8 % for AI, and 13.1 % and 99.8 % for human reading (p-value < 0.0001 and 0.9351). The expected reduction in LDCT reading workload when using AI as first reader was 74.7 %.

CONCLUSION: AI reading showed comparable sensitivity but lower specificity than human reading. High NPV of AI may support its use as a first reader in LCS.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Ledda RE, Valsecchi C, Sabia F, et al. Potential for AI as first reader in lung cancer screening. Eur J Radiol. 2025. PMID: 41308576. doi:10.1016/j.ejrad.2025.112561

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。