From the Health Innovation Big Data Center, Asan Institute for Life Science, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (H.L.).
Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.).
Stroke. 2020 Mar;51(3):860-866. doi: 10.1161/STROKEAHA.119.027611. Epub 2020 Jan 28.
Background and Purpose- We aimed to investigate the ability of machine learning (ML) techniques analyzing diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging to identify patients within the recommended time window for thrombolysis. Methods- We analyzed DWI and FLAIR images of consecutive patients with acute ischemic stroke within 24 hours of clear symptom onset by applying automatic image processing approaches. These processes included infarct segmentation, DWI, and FLAIR imaging registration and image feature extraction. A total of 89 vector features from each image sequence were captured and used in the ML. Three ML models were developed to estimate stroke onset time for binary classification (≤4.5 hours): logistic regression, support vector machine, and random forest. To evaluate the performance of ML models, the sensitivity and specificity for identifying patients within 4.5 hours were compared with the sensitivity and specificity of human readings of DWI-FLAIR mismatch. Results- Data from a total of 355 patients were analyzed. DWI-FLAIR mismatch from human readings identified patients within 4.5 hours of symptom onset with 48.5% sensitivity and 91.3% specificity. ML algorithms had significantly greater sensitivities than human readers (75.8% for logistic regression, =0.020; 72.7% for support vector machine, =0.033; 75.8% for random forest, =0.013) in detecting patients within 4.5 hours, but their specificities were comparable (82.6% for logistic regression, =0.157; 82.6% for support vector machine, =0.157; 82.6% for random forest, =0.157). Conclusions- ML algorithms using multiple magnetic resonance imaging features were feasible even more sensitive than human readings in identifying patients with stroke within the time window for acute thrombolysis.
背景与目的- 我们旨在研究机器学习（ML）技术分析弥散加权成像（DWI）和液体衰减反转恢复（FLAIR）磁共振成像的能力，以识别溶栓治疗推荐时间窗内的患者。方法- 通过应用自动图像处理方法，对发病 24 小时内的连续急性缺血性脑卒中患者的 DWI 和 FLAIR 图像进行分析。这些过程包括梗死分割、DWI 和 FLAIR 成像配准以及图像特征提取。从每个图像序列中捕获了总共 89 个向量特征，并用于 ML 中。为了对 ML 模型进行评估，将用于二进制分类（≤4.5 小时）的模型的灵敏度和特异性与 DWI-FLAIR 不匹配的人工读数的灵敏度和特异性进行了比较。结果- 共分析了 355 例患者的数据。DWI-FLAIR 不匹配的人工读数可识别出发病 4.5 小时内的患者，其灵敏度为 48.5%，特异性为 91.3%。ML 算法的灵敏度明显高于人工读数（逻辑回归为 75.8%，=0.020；支持向量机为 72.7%，=0.033；随机森林为 75.8%，=0.013），可检测出 4.5 小时内的患者，但特异性相当（逻辑回归为 82.6%，=0.157；支持向量机为 82.6%，=0.157；随机森林为 82.6%，=0.157）。结论- 使用多个磁共振成像特征的 ML 算法甚至比人工读数更灵敏，可在急性溶栓治疗时间窗内识别出卒中患者。
