University Institute of Diagnostic and Interventional Neuroradiology, Bern University Hospital, Inselspital (A.H., R.W.), University of Bern, Switzerland.
Stanford Stroke Center, Palo Alto, CA (S.C., M.G.L.).
Stroke. 2021 Jul;52(7):2328-2337. doi: 10.1161/STROKEAHA.120.030696. Epub 2021 May 7.
The ISLES challenge (Ischemic Stroke Lesion Segmentation) enables globally diverse teams to compete to develop advanced tools for stroke lesion analysis with machine learning. Detection of irreversibly damaged tissue on computed tomography perfusion (CTP) is often necessary to determine eligibility for late-time-window thrombectomy. Therefore, the aim of ISLES-2018 was to segment infarcted tissue on CTP based on diffusion-weighted imaging as a reference standard.
The data, from 4 centers, consisted of 103 cases of acute anterior circulation large artery occlusion stroke who underwent diffusion-weighted imaging rapidly after CTP. Diffusion-weighted imaging lesion segmentation was performed manually and acted as a reference standard. The data were separated into 63 cases for training and 40 for testing, upon which quality metrics (dice score coefficient, Hausdorff distance, absolute lesion volume difference, etc) were computed to rank methods based on their overall performance.
Twenty-four different teams participated in the challenge. Median time to CTP was 185 minutes (interquartile range, 180-238), the time between CTP and magnetic resonance imaging was 36 minutes (interquartile range, 25-79), and the median infarct lesion size was 15.2 mL (interquartile range, 5.7-45). The best performance for Dice score coefficient and absolute volume difference were 0.51 and 10.1 mL, respectively, from different teams. Based on the ranking criteria, the top team's algorithm demonstrated for average Dice score coefficient and average absolute volume difference 0.51 and 10.2 mL, respectively, outperforming the conventional threshold-based method (dice score coefficient, 0.3; volume difference, 15.3). Diverse algorithms were used, almost all based on deep learning, with top-ranked approaches making use of the raw perfusion data as well as methods to synthetically generate complementary information to boost prediction performance.
Machine learning methods may predict infarcted tissue from CTP with improved accuracy compared with threshold-based methods used in clinical routine. This dataset will remain public and can be used to test improvement in algorithms over time.
ISLES 挑战赛（脑梗死灶分割）使来自全球不同地区的团队能够通过机器学习竞相开发用于卒中病灶分析的先进工具。在确定是否适合进行晚期时间窗取栓术时，通常需要在 CT 灌注（CTP）上检测不可逆损伤组织。因此，ISLES-2018 的目标是基于弥散加权成像（DWI）作为参考标准，对 CTP 上的梗死组织进行分割。
该研究的数据来自 4 个中心的 103 例急性前循环大血管闭塞性卒中患者，这些患者在 CTP 后迅速进行 DWI 检查。手动进行 DWI 病变分割，并作为参考标准。数据分为 63 例用于训练，40 例用于测试，然后计算质量指标（Dice 评分系数、Hausdorff 距离、绝对病变体积差异等），根据整体性能对方法进行排名。
共有 24 个不同的团队参加了挑战赛。CTP 的中位时间为 185 分钟（四分位距，180-238），CTP 与磁共振成像之间的时间为 36 分钟（四分位距，25-79），中位梗死病变大小为 15.2ml（四分位距，5.7-45）。不同团队的 Dice 评分系数和绝对体积差异的最佳性能分别为 0.51 和 10.1ml。根据排名标准，排名最高的团队的算法在平均 Dice 评分系数和平均绝对体积差异方面的表现分别为 0.51 和 10.2ml，优于传统的基于阈值的方法（Dice 评分系数为 0.3，体积差异为 15.3ml）。使用了多种算法，几乎都是基于深度学习的，排名靠前的方法利用了原始灌注数据以及生成补充信息的方法，以提高预测性能。
与临床常规中使用的基于阈值的方法相比，机器学习方法可能更准确地预测 CTP 上的梗死组织。该数据集将保持公开，可用于测试算法随时间的改进。
