Performance of models highly depend not only on the used algorithm but also the data set it was applied to. This makes the comparison of newly developed tools to previously published approaches difficult. Either researchers need to implement others' algorithms first, to establish an adequate benchmark on their data, or a direct comparison of new and old techniques is infeasible. The Ischemic Stroke Lesion Segmentation (ISLES) challenge, which has ran now consecutively for 3 years, aims to address this problem of comparability. ISLES 2016 and 2017 focused on lesion outcome prediction after ischemic stroke: By providing a uniformly pre-processed data set, researchers from all over the world could apply their algorithm directly. A total of nine teams participated in ISLES 2015, and 15 teams participated in ISLES 2016. Their performance was evaluated in a fair and transparent way to identify the state-of-the-art among all submissions. Top ranked teams almost always employed deep learning tools, which were predominately convolutional neural networks (CNNs). Despite the great efforts, lesion outcome prediction persists challenging. The annotated data set remains publicly available and new approaches can be compared directly via the online evaluation system, serving as a continuing benchmark (www.isles-challenge.org).

ImportanceInternational guidelines recommend avoiding intravenous thrombolysis (IVT) in patients with ischemic stroke who have a recent intake of a direct oral anticoagulant (DOAC).ObjectiveTo determine the risk of symptomatic intracranial hemorrhage (sICH) associated with use of IVT in patients with recent DOAC ingestion.Design, Setting, and ParticipantsThis international, multicenter, retrospective cohort study included 64 primary and comprehensive stroke centers across Europe, Asia, Australia, and New Zealand. Consecutive adult patients with ischemic stroke who received IVT (both with and without thrombectomy) were included. Patients whose last known DOAC ingestion was more than 48 hours before stroke onset were excluded. A total of 832 patients with recent DOAC use were compared with 32 375 controls without recent DOAC use. Data were collected from January 2008 to December 2021.ExposuresPrior DOAC therapy (confirmed last ingestion within 48 hours prior to IVT) compared with no prior oral anticoagulation.Main Outcomes and MeasuresThe main outcome was sICH within 36 hours after IVT, defined as worsening of at least 4 points on the National Institutes of Health Stroke Scale and attributed to radiologically evident intracranial hemorrhage. Outcomes were compared according to different selection strategies (DOAC-level measurements, DOAC reversal treatment, IVT with neither DOAC-level measurement nor idarucizumab). The association of sICH with DOAC plasma levels and very recent ingestions was explored in sensitivity analyses.ResultsOf 33 207 included patients, 14 458 (43.5%) were female, and the median (IQR) age was 73 (62-80) years. The median (IQR) National Institutes of Health Stroke Scale score was 9 (5-16). Of the 832 patients taking DOAC, 252 (30.3%) received DOAC reversal before IVT (all idarucizumab), 225 (27.0%) had DOAC-level measurements, and 355 (42.7%) received IVT without measuring DOAC plasma levels or reversal treatment. The unadjusted rate of sICH was 2.5% (95% CI, 1.6-3.8) in patients taking DOACs compared with 4.1% (95% CI, 3.9-4.4) in control patients using no anticoagulants. Recent DOAC ingestion was associated with lower odds of sICH after IVT compared with no anticoagulation (adjusted odds ratio, 0.57; 95% CI, 0.36-0.92). This finding was consistent among the different selection strategies and in sensitivity analyses of patients with detectable plasma levels or very recent ingestion.Conclusions and RelevanceIn this study, there was insufficient evidence of excess harm associated with off-label IVT in selected patients after ischemic stroke with recent DOAC ingestion.

Cerebral arteries may exhibit a wide range of variation from normal anatomy, which can be incidentally discovered during imaging. Knowledge of such variants is crucial to differentiate them from pathologies, to understand the etiology of certain pathologies directly related to a vascular variant, and to depict the changes in collateral circulation in patients with certain variants. Detection of particular variants may lead to the discovery of other nonvascular or vascular anomalies, especially aneurysms, and may also affect planning of endovascular or neurosurgical interventions. In this review, we summarize the variants and anomalies of cerebral arteries seen on cross-sectional imaging classified by a morphological approach and categorize their significance from a clinical perspective. This structured review is intended to serve as a guide for daily use in clinical practice.