在中型急性国家卫生服务信托医院使用Brainomix e-CTA软件的真实世界临床经验
单中心回顾性服务评估项目分析2022年1月3日至2023年1月3日584例颅内CTA数据集,e-CTA检出大血管闭塞的敏感度0.78、特异度0.93、阳性预测值0.5、阴性预测值0.98,错误率9%。作者认为真实世界连续病例中e-CTA阳性预测值较差,主张AI软件临床使用前应进行本地验证。
要点
- e-CTA阳性预测值仅0.5(95%CI 0.42-0.58),假阳性常见原因为解剖误判和其他病变。
- e-CTA阴性预测值0.98(95%CI 0.96-0.99),假阴性常见于终末段颈内动脉闭塞。
- 作者主张AI软件临床使用前应进行本地验证。
核心结论
- 在真实世界NHS连续病例中e-CTA的阳性预测值较差。
- 作者主张AI软件在临床使用前应进行本地验证。
- 常见AI错误原因为解剖误判、存在其他病变以及终末段颈内动脉闭塞误判。
研究设计与样本
- 单中心回顾性服务评估项目,数据为2022年1月3日至2023年1月3日送至Brainomix的中型急性NHS信托医院数据。
- 共分析584例颅内CTA数据集,由e-CTA检测大血管闭塞。
- 参考标准由具有神经放射学专科培训的顾问放射科医生结合后续影像和临床记录确定。
诊断标准与关键数值
- e-CTA检测大血管闭塞敏感度0.78(95%CI 0.64-0.88)、特异度0.93(95%CI 0.9-0.95)、阳性预测值0.5(95%CI 0.42-0.58)、阴性预测值0.98(95%CI 0.96-0.99)。
- 584例中50例(9%)存在大血管闭塞。
- e-CTA错误率9%(52/584)。
- 假阳性常见原因:解剖误判21%(8/39)、其他病变13%(5/39)、无法分类39%(15/39)。
- 假阴性常见原因:终末段颈内动脉大血管闭塞55%(6/11)、无法分类18%(2/11)。
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: Artificial intelligence (AI) software including Brainomix "e-CTA" which detect large vessel occlusions (LVO) have clinical potential. We hypothesized that in real world use where prevalence is low, its clinical utility may be overstated.
METHODS: In this single centre retrospective service evaluation project, data sent to Brainomix from a medium size acute National Health Service (NHS) Trust hospital between January 3, 2022 and January 3, 2023 was reviewed. 584 intracranial computed tomography angiogram (CTA) datasets were analysed for LVO by e-CTA. The e-CTA output and radiology report were compared to ground truth, defined by a consultant radiologist with fellowship neuroradiology training, with access to subsequent imaging and clinical notes. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated.
RESULTS: Of 584 cases (45% female, mean age 70 ± 16 years), 9% (n = 50) had LVO. e-CTA had a sensitivity of 0.78 (95% CI 0.64-0.88), specificity of 0.93 (0.9-0.95), PPV of 0.5 (0.42-0.58), and NPV of 0.98 (0.96-0.99). e-CTA had an error rate of 9% (52/584). Erroneous cases were categorized into causes for error. Common causes for false positives included incorrect anatomy (21%, 8/39) and other pathology (13%, 5/39), with several uncategorizable cases (39%, 15/39). Common causes for false negatives included LVO within the terminal internal carotid artery (ICA) (55%, 6/11) and uncategorizable (18%, 2/11).
CONCLUSIONS: We demonstrated that PPV of e-CTA is poor in consecutive cases in a real-world NHS setting. We advocate for local validation of AI software prior to clinical use.
ADVANCES IN KNOWLEDGE: Common AI errors were due to anatomical misidentification, presence of other pathology, and misidentifying LVO in the terminal ICA.
引用
- Merchant F, Choulerton J, James R, et al. Real world clinical experience of using Brainomix e-CTA software in a medium size acute National Health Service Trust. Br J Radiol. 2025. PMID: 39878893. doi:10.1093/bjr/tqaf019
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。