Background Flow diverters have been widely used in clinical practice for more than a decade. However, most outcome data are limited to 1 year timepoints. This study aims to offer meta-analysis data on long-term (>1 year) safety and effectiveness results for patients with aneurysms treated with flow diverters. Methods PubMed, Web of Science, Embase, and SCOPUS were searched up to February 24, 2022 using the AutoLit platform. We included primary studies assessing the long-term outcomes for flow diverter devices to manage unruptured internal carotid artery aneurysms with a follow-up period of >1 year. The meta-analysis was carried out using Comprehensive Meta-Analysis software (CMA). Results Eleven studies were included in the meta-analysis. The pooled occlusion rates after flow diversion treatment for unruptured intracranial brain aneurysms were 77%, 87.4%, 84.5%, 89.4%, 96% for 1 year, 1–2 years, 2 years, 3 years, and 5 years follow-up, respectively. The in-stent stenosis rate was 4.8% and the retreatment rate for the long-term follow-up period was 5%. No delayed rupture of the aneurysm was reported, and there was one case of delayed ischemic stroke. The sensitivity analysis of the prospective studies showed a complete occlusion rate of 83.5% and 85.2% for 1 and 3 years of follow-up, respectively. Conclusion Flow diverters are safe and effective in short- and long-term follow-up and rarely cause serious delayed side effects.

Background and Purpose: Mechanical thrombectomy greatly improves stroke outcomes. Nonetheless, some patients fall short of full recovery despite good reperfusion. The purpose of this study was to develop machine learning (ML) models for the pre-interventional prediction of functional outcome at 3 months of thrombectomy in acute ischemic stroke (AIS), using clinical and auto-extractable radiological information consistently available upon first emergency evaluation. Materials and Methods: A two-center retrospective cohort of 293 patients with AIS who underwent thrombectomy was analyzed. ML models were developed to predict dichotomized modified Rankin score at 90 days (mRS-90) using clinical and imaging features, both separately and combined. Conventional and experimental imaging biomarkers were quantified using automated image-processing software from non-contract computed tomography (CT) and computed tomography angiography (CTA). Shapley Additive Explanation (SHAP) was applied for model interpretability and predictor importance analysis of the optimal model. Results: Merging clinical and imaging features returned the best results for mRS-90 prediction. The best performing classifier was Extreme Gradient Boosting (XGB) with an area under the receiver operating characteristic curve (AUC) = 84% using selected features. The most important classifying features were age, baseline National Institutes of Health Stroke Scale (NIHSS), occlusion side, degree of brain atrophy [primarily represented by cortical cerebrospinal fluid (CSF) volume and lateral ventricle volume], early ischemic core [primarily represented by e-Alberta Stroke Program Early CT Score (ASPECTS)], and collateral circulation deficit volume on CTA. Conclusion: Machine learning that is applied to quantifiable image features from CT and CTA alongside basic clinical characteristics constitutes a promising automated method in the pre-interventional prediction of stroke prognosis. Interpretable models allow for exploring which initial features contribute the most to post-thrombectomy outcome prediction overall and for each individual patient outcome.

The best-performing open-weight language models had high accuracy for automated extraction of structured clinical data points from unstructured radiology and pathology reports, with retrieval-augmented generation improving performance for complex reports.

Background Successful recanalization after endovascular thrombectomy serves as the primary endpoint in clinical trials and is a crucial predictor of long-term outcomes. Radiographic outcomes for various interventions have been shown to vary based on the type of interpreter, including the site interventionalist compared with an independent reader. Objective To compare angiographic outcomes in stroke thrombectomy procedures based on the type of reader. Methods A systematic literature search was conducted in Medline, EMBASE, Scopus, and Web-of-Science through February 2022. We included primary studies that reported core laboratory-read and operator angiographic outcomes after mechanical thrombectomy for ischemic stroke. Furthermore, study-defined successful recanalization data were collected. Results Eight studies were included with 4797 patients, 51.2% of whom were male. Four thousand, four hundred and thirty-one patients had core readings, and 4211 had operator readings. Study-defined successful recanalization was significantly higher for operator (84%, 3543/4211) examinations than for core laboratory-read (78.4%, 3476/4431) examinations (p<0.001; OR=1.462, 95% CI 1.175 to 1.819). The modified Thrombolysis in Cerebral Infarction (mTICI) scale score of ≥2 b was higher for operator (85%, 3341/3929) examinations than for core laboratory-read (78.6%, 3107/3952) examinations (p<0.001; OR=1.349, 95% CI 1.071 to 1.701). mTICI 3 was significantly higher for operator (54.6%, 1000/1832) examinations than for core laboratory-read (39.9%, 731/1832) examinations (p<0.001; OR=1.823, 95% CI 1.598 to 2.081). Conclusion Operator angiographic reads are statistically significantly higher than core laboratory-read readings following stroke thrombectomy, especially for complete recanalization. These differences should be considered when interpreting reports of angiographic outcomes after thrombectomy.

Background: Computed tomography (CT) angiography collateral score (CTA-CS) is an important clinical outcome predictor following mechanical thrombectomy for ischemic stroke with large vessel occlusion (LVO). The present multireader study aimed to evaluate the performance of e-CTA software for automated assistance in CTA-CS scoring. Materials and Methods: Brain CTA images of 56 patients with anterior LVO were retrospectively processed. Twelve readers of various clinical training, including junior neuroradiologists, senior neuroradiologists, and neurologists graded collateral flow using visual CTA-CS scale in two sessions separated by a washout period. Reference standard was the consensus of three expert readers. Duration of reading time, inter-rater reliability, and statistical comparison of readers’ performance metrics were analyzed between the e-CTA assisted and unassisted sessions. Results: e-CTA assistance resulted in significant increase in mean accuracy (58.6% to 67.5%, p = 0.003), mean F1 score (0.574 to 0.676, p = 0.002), mean precision (58.8% to 68%, p = 0.007), and mean recall (58.7% to 69.9%, p = 0.002), especially with slight filling deficit (CTA-CS 2 and 3). Mean reading time was reduced across all readers (103.4 to 59.7 s, p = 0.001), and inter-rater agreement in CTA-CS assessment was increased (Krippendorff's alpha 0.366 to 0.676). Optimized occlusion laterality detection was also noted with mean accuracy (92.9% to 96.8%, p = 0.009). Conclusion: Automated assistance for CTA-CS using e-CTA software provided helpful decision support for readers in terms of improving scoring accuracy and reading efficiency for physicians with a range of experience and training backgrounds and leading to significant improvements in inter-rater agreement.