Development and validation of an explainable machine learning prediction model for futile recanalization after mechanical thrombectomy in acute large vessel occlusion stroke

Journal of NeuroInterventional Surgery, Journal Year: 2025, Volume and Issue: unknown, P. jnis - 023106

Published: April 11, 2025

Background Mechanical thrombectomy (MT) is the primary treatment for acute ischemic stroke (AIS) caused by large vessel occlusion (LVO). However, likelihood of futile recanalization (FR) at 90 days post-MT remains high. Methods This study included 534 AIS patients with anterior circulation LVO who underwent MT, outcome being FR. The derivation cohort consisted 445 (June 2018–June 2023), while temporal validation had 89 (July 2023–June 2024). was split into 70% training and 30% internal sets. Eleven machine learning (ML) models were trained, tested, compared, best-performing model selected optimization validation. SHapley Additive exPlanations (SHAP) used interpretation. Results CatBoost showed best discriminative ability among 11 ML models. After feature selection dimensionality reduction, a final explainable 12 features established, accurately predicting FR in both (area under curve (AUC)=0.915) (AUC=0.930) validations. has been deployed as web application clinical use. Conclusion We developed prediction key that demonstrates excellent performance deployment this offers promising tool clinicians to assess risk, potentially enhancing patient improving personalized care.

Language: Английский

---

Evolucollateral dynamics in stroke: Evolutionary pathophysiology, remodelling and emerging therapeutic strategies

European Journal of Neuroscience, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 5, 2024

Abstract Leptomeningeal collaterals (LMCs) are crucial in mitigating the impact of acute ischemic stroke (AIS) by providing alternate blood flow routes when primary arteries obstructed. This article explores evolutionary pathophysiology LMCs, highlighting their critical function and genetic molecular mechanisms governing development remodelling. We address translational challenges applying animal model findings to human clinical scenarios, emphasizing need for further research validate emerging therapies—such as pharmacological agents, gene therapy mechanical interventions—in settings, aimed at enhancing collateral perfusion. Computational modelling emerges a promising method integrating experimental data, which requires precise parameterization empirical validation. introduce ‘Evolucollateral Dynamics’ hypothesis, proposing novel framework that incorporates biology principles into therapeutic strategies, offering new perspectives on circulation. hypothesis emphasizes role predispositions environmental influences circulation, may strategies optimize treatment outcomes. Future must incorporate data create robust protocols, thereby maximizing potential LMCs improving outcomes patients.

Language: Английский

---

Evaluating Machine Learning Models for Stroke Prognosis and Prediction in Atrial Fibrillation Patients: A Comprehensive Meta-Analysis

Diagnostics, Journal Year: 2024, Volume and Issue: 14(21), P. 2391 - 2391

Published: Oct. 26, 2024

Atrial fibrillation (AF) complicates the management of acute ischemic stroke (AIS), necessitating precise predictive models to enhance clinical outcomes. This meta-analysis evaluates efficacy machine learning (ML) in three key areas: prognosis AF patients, prediction and patients. The study aims assess accuracy variability ML forecasting AIS outcomes detecting while exploring benefits limitations integrating these into practice.

Language: Английский

---

Involvement of Subinsular Territory Stroke as Predictor of Outcome after Successful Endovascular Recanalization of Left Middle Cerebral Artery Occlusion

Brain Sciences, Journal Year: 2024, Volume and Issue: 14(9), P. 885 - 885

Published: Aug. 30, 2024

Subinsular stroke (subIS) can occur between the penetrating middle cerebral artery (MCA) branches, which clinical and radiological findings sometimes encounter in patients after recanalization of left proximal MCA occlusion. However, no supportive data are available to clarify this relationship. This study investigated whether involvement subIS impact outcomes successful reperfusion therapy. Data from 152 consecutive who underwent endovascular thrombectomy 2019 2023 were collected. A 3-month functional independence defined as a modified Rankin Scale 0-2 (primary outcome) influencing factors analyzed retrospectively. Recanalization was achieved 35 patients, whom 11 (31%) developed subIS. Patients with older age (81 vs. 75;

Language: Английский

---

The need for careful consideration of futile recanalization in acute ischemic stroke patients undergoing endovascular treatment

Journal of the Neurological Sciences, Journal Year: 2024, Volume and Issue: 462, P. 123104 - 123104

Published: June 20, 2024

Language: Английский

---