Beyond Conventional Imaging: From Time-Based to Data-Driven Decision-Making in Acute Ischemic Stroke

Reperfusion therapy has profoundly transformed acute ischemic stroke (AIS) care. Initially, treatment decisions relied primarily on time from symptom onset. However, growing evidence has shown that the extent of irreversibly injured tissue and the presence of salvageable brain tissue determine the efficacy and safety of reperfusion therapy. This caused a paradigm shift from time-based selection toward tissue-based assessment, placing neuroimaging at the center of clinical decision-making. This narrative review traces the evolution of imaging-based stroke assessment in the reperfusion era. We review the development and clinical impact of the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) and diffusion-weighted imaging (DWI)-ASPECTS, which translated the pathophysiological concept of the ischemic core into a practical, reproducible, routine clinical tool. We discuss the emergence of imaging and clinical mismatch concepts, including perfusion-based ischemic core-penumbra mismatch, as well as clinical-diffusion, MRA-diffusion, and DWI-FLAIR mismatches, which address the limitations of strict time-based criteria and expand treatment opportunities, particularly in patients with unknown onset time. Finally, we examine the growing role of artificial intelligence (AI)-driven automated imaging platforms in acute stroke care, enabling rapid, standardized, and quantitative assessment of ischemic core, penumbra, and vascular pathology. These tools support clinical decision-making under severe time constraints and reduce inter-reader and inter-institutional variability. This iterative process of trial and refinement has shaped the current framework of tissue-based, data-driven stroke care. We discuss how integrating established imaging concepts with emerging AI technologies may further advance precision reperfusion therapy and improve equity and outcomes in AIS treatment.