Introduction: While the machine learning (ML) model's black-box nature presents a significant barrier to effective clinical application, the dynamic nature of stroke patients' recovery further undermines the reliability of established predictive scores and models, making them less suitable for accurate prediction and appropriate patient care. This research is aimed at building and evaluating an interpretable ML-based model, which would perform outcome prediction at different time points of patients' recovery, giving more secure and understandable output through interpretable packages. Materials and Methods: A retrospective analysis was conducted on acute ischemic stroke (AIS) patients treated with alteplase at the Neurology Clinic of the University Clinical Center of Vojvodina (Novi Sad, Serbia), for 14 years. Clinical data were grouped into four categories based on collection time-baseline, 2-h, 24-h, and discharge features-serving as inputs for three different classifiers-support vector machine (SVM), logistic regression (LR), and random forest (RF). The 90-day modified Rankin scale (mRS) was used as the outcome measure, distinguishing between favorable (mRS ≤ 2) and unfavorable outcomes (mRS ≥ 3). Results: The sample was described with 49 features and included 355 patients, with a median age of 67 years (interquartile range (IQR) 60-74 years), 66% being male. The models achieved strong discrimination in the testing set, with area under the curve (AUC) values ranging from 0.80 to 0.96. Additionally, they were compared with a model based on the DRAGON score, which showed an AUC of 0.760 (95% confidence interval (CI), 0.640-0.862). The decision-making process was more thoroughly understood using interpretable packages: Shapley additive explanation (SHAP) and local interpretable model-agnostic explanation (LIME). They revealed the most significant features at both the group and individual patient levels. Conclusions and Clinical Implications: This study demonstrated the moderate to strong efficacy of interpretable ML-based models in predicting the functional outcomes of alteplase-treated AIS patients. In all constructed models, age, onset-to-treatment time, and platelet count were recognized as the important predictors, followed by clinical parameters measured at different time points, such as the National Institutes of Health Stroke Scale (NIHSS) and systolic and diastolic blood pressure values. The dynamic approach, coupled with interpretable models, can aid in providing insights into the potential factors that could be modified and thus contribute to a better outcome.

Background and Aims: Histological analysis of thrombi can enhance the understanding of pathophysiology. We aimed to analyze EVT-retrieved thrombi in cerebral venous thrombosis (CVT), compare them with acute ischemic stroke (AIS) thrombi, and correlate their composition with CT density.Methods: Retrospective case-series, including five CVT and 10 AIS cases treated with EVT. Thrombus sections were stained with hematoxylin and eosin; Picro Mallory for RBCs, fibrin, and collagen; and Prussian Blue for iron plus immunohistochemical staining with anti-CD61 (platelets), anti-MPO (neutrophils), anti-CD3 (T-cells), anti-CD20 (B-cells), anti-CD34 (endothelial cells), anti-CD68 (macrophages), and anti-citH3 (NETs). Thrombus components were quantified (Orbit) and expressed as a percentage of total area. The CVT-thrombus relative density (rHU) was calculated as HU thrombus/HU contralateral.Results: All CVT cases showed extensive thrombosis. Four patients had prior anticoagulation, and four had rHU > 1.00 with CT hyperdensity. The etiologies were heterogeneous. CVT thrombi were rich in red blood cells and displayed variable histological features, including signs of early organization. Compared to arterial thrombi, venous thrombi exhibited larger size (surface area 185.6 mm2 [IQR 83.0–237.9] vs. 21.8 mm2 [IQR 8.8–77.8]; p = 0.028) and lower fibrin content (16.6% [IQR 13.9–31.5] vs. 46.5% [IQR 25.1–49.5]; p = 0.036), with no other significant differences in composition. Low fibrin content and high RBC-to-fibrin ratio (R −0.9 and R 0.9, respectively; p = 0.047 for both) showed a significant correlation with rHU.Conclusion: Our exploratory study first shows that CVT thrombi are larger than AIS thrombi, with higher RBC content and lower fibrin, matching CT density. These findings enhance the understanding of CVT pathophysiology but need validation.

Background: Calcified cerebral emboli (CCEs) represent a rare cause of acute ischemic stroke and can pose technical challenges for neurointerventionalists. The few studies on endovascular thrombectomy (EVT) of CCE to date show poor recanalization rates and unfavorable outcomes.Objective: This study is aimed at investigating the technical and clinical results concerning EVT of CCE compared with noncalcified cerebral emboli (NCCEs).Methods: All cases of EVT for acute stroke from January 2014 to December 2021 from a single center were analyzed retrospectively. Emboli with a maximum density of ≥ 130 Hounsfield units on preinterventional CT scans were considered calcified. Propensity score matching was performed to compare technical and clinical results between patients with CCE and NCCE.Results: CCEs were present in 26 of 1004 cases (2.6%). Successful recanalization (mTICI ≥ 2b) was achieved less frequently in CCE (CCE: 62%, NCCE: 92%, p = 0.009). Also, first-pass reperfusion was less common in CCE (CCE: 12%, NCCE: 46%, p = 0.006). In CCE, infarct growth was more frequent (CCE: 81%, NCCE: 42%, p = 0.004) and more severe (p = 0.005). National Institutes of Health Stroke Scale improvement after EVT was lower in CCE patients (CCE: median 2, range −23 to 20, interquartile range (IQR) 2.75; NCCE: median 5, range −8 to 17, IQR 11, p = 0.008).Conclusion: First-pass reperfusion is less common in EVT of CCE. Also, there is a more frequent and severe infarct growth in CCE patients after EVT, which helps to understand the poorer clinical results. Thrombectomy devices optimized for CCE are desirable to improve outcomes in this subgroup of stroke patients.

Over 80% of stroke survivors experience walking dysfunction, impacting quality of life. Rehabilitation is crucial for gait recovery, and accurate assessments facilitate tailored programs. While computerized gait analysis is the gold standard, it is costly and requires specialized training, making observational gait analysis (OGA) more common. However, OGA can also be time‐consuming. This study validates the Malta Gait Scale (MGS), a concise, illustrated 7‐item observational tool using video recordings for gait measurements. The aim is to provide an effective, time‐efficient method for gait evaluations by comparing the MGS with the established Wisconsin Gait Scale (WGS) and Gait Assessment Intervention Tool (GAIT), which have 14 and 31 items, respectively. Forty‐nine participants were included in a retrospective study to validate the MGS. We evaluated its reliability using weighted Cohen’s kappa (κ) for intrarater and interrater reliability. Concurrent validity was assessed by comparing the MGS with the WGS and GAIT scales using Spearman’s rho (ρ). The Wilcoxon test assessed the efficacy of the MGS in detecting rehabilitation‐induced changes, differentiating healthy from stroke participants, and evaluating time efficiency. The MGS demonstrated almost perfect agreement, with interrater and intrarater κ values of 0.952 and 0.977, respectively. It showed high positive correlations with the WGS and GAIT, with ρ values of 0.898 and 0.877. MGS required an average administration time of 7 min and 29 s, significantly less than the WGS (27 min and 46 s) and GAIT (50 min and 6 s) (p < 0.001). Following rehabilitation, significant improvements were observed in patients using both the MGS and WGS scales (p = 0.018), and the MGS effectively distinguished between healthy individuals and stroke patients (p < 0.001). The MGS is a valid, reliable, and efficient tool for gait assessment in stroke survivors, supporting smartphone use and facilitating rapid measurements in clinical settings where time is critical.

Background: Cerebral venous thrombosis (CVT) is a rare but potentially debilitating form of stroke. Current management guidelines recommend a course of low molecular weight heparin (LMWH) followed by an oral vitamin K antagonist. However, there is an emerging body of evidence to support the use of direct oral anticoagulant (DOAC) medications. Here, we assess the efficacy of rivaroxaban compared to the standard of care in a resource‐limited setting.Methods: The study was designed as a Phase III, prospective, parallel, open‐label, randomized controlled trial conducted in three sites in Syria. Seventy‐one participants met inclusion criteria and were randomized 1:1 to receive either rivaroxaban or warfarin following initial bridging with LMWH for 3.5–12 days. The primary outcome was functional improvement determined by the Barthel Index. Secondary outcomes were adverse events during follow‐up, including CVT recurrence, thrombotic events, intracranial pressure (ICP) requiring shunt placement, extra and intracranial bleeding, neurological deficit, and all‐cause mortality.Results: Barthel Index scores did not differ between the study cohorts at 1‐, 2‐, 3‐, 4‐, 5‐, or 6‐month follow‐up. Secondary analysis yielded no difference in rates of adverse effects or return of CVT. Two patients in the warfarin group developed major extracranial bleeds (uterine bleeding); however, there were no other extracranial or intracranial bleeds or thrombotic events reported. Rates of all‐cause mortality and all assessed adverse effects were similar between the groups.Conclusion: We offer a prospective, parallel randomized controlled trial that suggests rivaroxaban may have comparable safety and efficacy when compared to warfarin for the treatment of CVT. Importantly, we offer the first randomized control trial of oral anticoagulants for the treatment of CVT in a resource‐limited setting, providing support for the evolving literature and suggesting the safety and efficacy of oral anticoagulants in the management of CVT.Trial Registration: ClinicalTrials.gov identifier: NCT04569279

Poststroke cognitive impairment (PSCI) refers to any level of cognitive decline occurring after a stroke, ranging from mild to severe impairments, while PSCI with dementia describes a more severe form where the cognitive decline significantly affects daily functioning and meets the clinical criteria for dementia. PSCI occurs in more than half of individuals who have had a stroke. Despite its high prevalence, the pharmacotherapeutic options for PSCI are limited. Several pharmacotherapeutic options like cholinesterase inhibitors (e.g., donepezil, galantamine, and rivastigmine) and N-methyl-d-aspartate receptor antagonists (e.g., memantine) have shown potential in improving cognitive functions. However, their overall effectiveness remains inconsistent across different studies and patient populations. Newer drugs such as citicoline, cilostazol, and antidepressants have shown promise, but further research is needed to validate their efficacy and safety specifically for PSCI management.

Background: Cognitive impairments and depression are common after stroke. Noninvasive treatments like transcranial direct current stimulation (tDCS) and mindfulness-based interventions have shown potential for improving these outcomes, though their effects on stroke survivors remain unclear. This study is aimed at evaluating the efficacy of mindfulness and tDCS in enhancing cognitive function and alleviating depression in stroke survivors. Methods: This randomized controlled trial, conducted from July 2021 to July 2022, included 30 stroke survivors divided into three groups: mindfulness (n = 5), tDCS (n = 14), and control (n = 11). Cognitive function was measured using Addenbrooke's Cognitive Examination-III (ACE-III), and depression was assessed using the Beck Depression Inventory-II (BDI-II) before and after interventions. The tDCS group received 10 sessions of anodal stimulation, and the mindfulness group underwent eight weekly sessions of mindfulness-based stress reduction. Data were analyzed using paired t-tests for within-group comparisons and ANOVA for between-group differences. Results: The tDCS group showed significant improvement in cognitive function, with ACE-III scores increasing by 9.14 ± 8.24 points (p = 0.02). Fluency and orientation scores also improved significantly in this group (p < 0.001 and p = 0.01, respectively). No significant cognitive changes were observed in the mindfulness group. Depression scores (BDI-II) did not change significantly in any group. Conclusions: tDCS significantly improved cognitive performance, particularly in fluency and orientation, while mindfulness showed no significant cognitive or depression-related effects. Future studies should explore the long-term impact of these interventions in stroke rehabilitation. Trial Registration: ClinicalTrials.gov identifier: IRCT20090716002195N3.

Introduction: Stroke is a leading global contributor to mortality and disability. Low- and middle-income countries are disproportionately affected and account for 87% of stroke-related disabilities and 70% of stroke-related fatalities. The challenges of stroke care accessibility in Africa are compounded by financial constraints, geographical barriers, and inadequate healthcare infrastructure, necessitating the adoption of innovative models such as telestroke. Telestroke is a critical component of modern stroke care systems. Telestroke enables real-time remote assessments, optimizes patient triage and hospital transfers, improves the efficiency of thrombolysis administration, and enhances poststroke management by mitigating logistical and mobility-related challenges. This demonstrates telestroke's potential to expand access to specialized stroke care, improve functional outcomes, and address critical gaps in stroke management within underserved regions such as Africa. This paper assesses the advantageous impact of telestroke on stroke management, with the aim of drawing global insights for Africa.Methodology: This scoping review adhered to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines. A comprehensive search was conducted across ProQuest, PubMed, Google Scholar, and Scopus to identify peer-reviewed studies published in English from 2017 to 2023. This ensured the inclusion of the most recent advancements in telestroke research.Results: The initial literature search retrieved 881 articles, of which 143 duplicates (16.2%) and 58 non-English studies (6.6%) were removed, followed by the exclusion of 451 nonpeer-reviewed publications (51.2%) and 128 articles (14.5%) unrelated to the study area, leaving 101 studies (11.5%) for full-text review. After further screening, 70 studies were excluded for not aligning with the study's title, objectives, or key search terms. This resulted in 31 studies (3.5%) being included in the final analysis, with 21 studies originating from outside Africa. The limited availability of high-indexed, peer-reviewed African telestroke studies highlighted a research gap, impacting the generalizability of findings.Conclusion: Telestroke has demonstrated significant benefits in stroke management, including improved functional outcomes, reduced onset-to-treatment time, enhanced diagnostic accuracy, and increased healthcare accessibility, particularly in medically underserved regions. However, its implementation in Africa faces challenges related to ethical concerns, technological infrastructure, regulatory inconsistencies, financial sustainability, and limited clinician buy-in. This necessitates strategic interventions such as standardized regulatory frameworks, network expansion, sustainable financing, capacity-building, and the integration of cost-effective imaging technologies to enhance stroke care delivery across the continent.

Stroke is a high morbidity and mortality disease that poses a serious threat to people's health. Early recognition of the various warning signs of stroke is necessary so that timely clinical intervention can help reduce the severity of stroke. Deep neural networks have powerful feature representation capabilities and can automatically learn discriminant features from large amounts of data. This paper uses a range of physiological characteristic parameters and collaborates with deep neural networks, such as the Wasserstein generative adversarial networks with gradient penalty and regression network, to construct a stroke prediction model. Firstly, to address the problem of imbalance between positive and negative samples in the stroke public data set, we performed positive sample data augmentation and utilized WGAN-GP to generate stroke data with high fidelity and used it for the training of the prediction network model. Then, the relationship between observable physiological characteristic parameters and the predicted risk of suffering a stroke was modeled as a nonlinear mapping transformation, and a stroke prediction model based on a deep regression network was designed. Finally, the proposed method is compared with commonly used machine learning-based classification algorithms such as decision tree, random forest, support vector machine, and artificial neural networks. The prediction results of the proposed method are optimal in the comprehensive measurement index F. Further ablation experiments also show that the designed prediction model has certain robustness and can effectively predict stroke diseases.