From the Department of Radiology, New York University Langone Medical Center, 660 First Ave, 2nd Floor, New York, NY 10016 (S.D.); Department of Neurology, Stanford University Hospital, Stanford, Calif (M.L., C.V., J.H.); Department of Neurology, Ente Ospedaliero Cantonale, Lugano, Switzerland (C.C.); Departments of Neurology (F.L., H.C., F.R.) and Radiology (F.M.), Hospital Geral de Fortaleza, Fortaleza, Brazil; Department of Neurology, Kansas University Medical Center, Kansas City, Kan (A.Q.); LifeBridge, Baltimore, Md (H.H.); and Boulder Statistics, Boulder, Colo (K.C.).
Radiology. 2021 Mar;298(3):665-670. doi: 10.1148/radiol.2021202734. Epub 2021 Jan 12.
Background Identification of large vessel occlusion (LVO) is critical to the management of acute ischemic stroke and prerequisite to endovascular therapy in recent trials. Increasing volumes and data complexity compel the development of fast, reliable, and automated tools for LVO detection to facilitate acute imaging triage. Purpose To investigate the performance of an anterior circulation LVO detection platform in a large mixed sample of individuals with and without LVO at cerebrovascular CT angiography (CTA). Materials and Methods In this retrospective analysis, CTA data from recent cerebrovascular trials (CRISP [ NCT01622517] and DASH) were enriched with local repositories from 11 worldwide sites to balance demographic and technical variables in LVO-positive and LVO-negative examinations. CTA findings were reviewed independently by two neuroradiologists from different institutions for intracranial internal carotid artery (ICA) or middle cerebral artery (MCA) M1 LVO; these observers were blinded to all clinical variables and outcomes. An automated analysis platform was developed and tested for prediction of LVO presence and location relative to reader consensus. Discordance between readers with respect to LVO presence or location was adjudicated by a blinded tertiary reader at a third institution. Sensitivity, specificity, and receiver operating characteristics were assessed by an independent statistician, and subgroup analyses were conducted. Prespecified performance thresholds were set at a lower bound of the 95% CI of sensitivity and specificity of 0.8 or greater at mean times to notification of less than 3.5 minutes. Results A total of 217 study participants (mean age, 64 years ± 16 [standard deviation]; 116 men; 109 with positive findings of LVO) were evaluated. Prespecified performance thresholds were exceeded (sensitivity, 105 of 109 [96%; 95% CI: 91, 99]; specificity, 106 of 108 [98%; 95% CI: 94, 100]). Sensitivity and specificity estimates across age, sex, location, and vendor subgroups exceeded 90%. The area under the receiver operating characteristic curve was 99% (95% CI: 97, 100). Mean processing and notification time was 3 minutes 18 seconds. Conclusion The results confirm the feasibility of fast automated high-performance detection of intracranial internal carotid artery and middle cerebral artery M1 occlusions. © RSNA, 2021 See also the editorial by Kloska in this issue.
背景 明确大血管闭塞（LVO）对于急性缺血性脑卒中的治疗至关重要，也是近期血管内治疗试验的前提。由于数据量不断增加且日趋复杂，因此需要开发快速、可靠和自动化的 LVO 检测工具，以辅助急性影像分诊。目的 旨在探讨一种在前循环 LVO 检测平台在伴有和不伴有脑血管 CT 血管造影（CTA） LVO 的大量混合个体中的性能。材料与方法 本回顾性分析中，对来自近期脑血管试验（CRISP [NCT01622517]和 DASH）的 CTA 数据进行了扩充，纳入了来自 11 个全球站点的本地存储库，以平衡 LVO 阳性和 LVO 阴性检查的人口统计学和技术变量。由两名来自不同机构的神经放射科医生独立对颅内颈内动脉（ICA）或大脑中动脉（MCA）M1 LVO 进行 CTA 结果评估；这些观察者对所有临床变量和结果均不知情。开发了一种自动分析平台，并针对预测 LVO 存在和位置与读者共识进行了测试。在第三家机构，由一位盲法三级读者对读者之间关于 LVO 存在或位置的差异进行裁决。由独立统计学家评估观察者间的灵敏度、特异性和受试者工作特征曲线，并进行亚组分析。以灵敏度和特异性的 95%CI 下限 0.8 或更高为预设性能阈值，设定了通知时间小于 3.5 分钟的下限。结果 共评估了 217 名研究参与者（平均年龄，64 岁±16[标准差]；116 名男性；109 名 LVO 阳性）。预设性能阈值均得到满足（灵敏度为 109 例中的 105 例[96%；95%CI：91，99]；特异性为 108 例中的 106 例[98%；95%CI：94，100]）。年龄、性别、位置和供应商亚组的灵敏度和特异性估计值均超过 90%。受试者工作特征曲线下面积为 99%（95%CI：97，100）。平均处理和通知时间为 3 分 18 秒。结论 研究结果证实了快速自动高性能检测颅内颈内动脉和大脑中动脉 M1 闭塞的可行性。©RSNA，2021 本期还刊登了 Kloska 的评论文章。
