1389 A Meta-Analysis and Systematic Review of Artificial Intellgence Methods for Predicting Hemorrhagic Transformation Using Neuroimaging in Acute Ischemic Stroke

INTRODUCTION: Early and reliable prediction of hemorrhagic transformation (HT) in patients with acute ischemic stroke (AIS) is crucial for treatment decisions and early intervention. METHODS: A systematic search of PubMed, EMBASE, and Web of Science was conducted until February 19, 2024. Inclusion criteria were: patients with AIS; patients who received reperfusion therapy; AI/ML algorithm using any imaging modality applied to predict hemorrhagic transformation; presence of sufficient data on the detective performance of AI/ML algorithms. Exclusion criteria were: reviews, letters, guidelines, editorials, or errata; case reports or series with less than 20 patients; studies with overlapping cohorts; articles lacking algorithms that operate solely on images; articles not detailing the algorithm used. The quality of eligible studies was assessed using the QUADAS-2 and CLAIM. Pooled sensitivity, specificity, and diagnostic odds ratio (DOR) were calculated using a random-effects model, and a summary receiver operating characteristic curve was performed using the Reitsma method. RESULTS: We identified six eligible studies, which included 1,640 patients. Aside from an unclear risk of bias regarding flow and timing identified in two of the studies, all studies showed low risk of bias and applicability concerns in all categories. Pooled sensitivity, specificity, and DOR were 0.849, 0.878, and 45.598, respectively. CONCLUSIONS: AI/ML models can reliably predict the occurrence of HT in AIS patients. More prospective studies are needed for subgroup analyses and higher clinical certainty and usefulness.