Abstract
Background and Objectives: Acute ischemic stroke (AIS) caused by large vessel occlusion (LVO) remains a major cause of disability and mortality. Mechanical thrombectomy (MT) improves outcomes, but recovery varies. This study assessed the prognostic value of hypoperfusion intensity ratio (HIR), collateral circulation, and other clinical/imaging factors. Materials and Methods: This retrospective cohort study included 96 LVO patients treated with MT with or without intravenous thrombolysis (IVT) between 2020 and 2024 at a tertiary hospital. Inclusion required multimodal CT (CT, CTA, CTP) and clinical data (NIHSS, mRS). HIR, core volume, CBV index, mismatch ratio, and collateral status were evaluated using artificial intelligence (AI)-based software. Univariate/multivariate logistic regression identified predictors of poor outcome (mRS > 3 at 90 days). Results: Lower HIR (<0.5) and good collaterals were associated with favourable outcomes (p < 0.001). Multivariate analysis identified HIR, initial NIHSS, and procedure duration as independent predictors of poor outcome. CTP-derived core volume, cerebral blood volume index, and mismatch ratio were also significant predictors. ROC analysis showed the highest AUC for core volume (0.810). Diabetes mellitus was associated with a worse prognosis compared to other clinical factors. Conclusions: HIR and collateral status are independent predictors of functional recovery after MT. CTP-derived core volume and CBV index have strong prognostic value. AI-based perfusion analysis supports patient selection and risk stratification.