Beyond one-size-fits-all in acute ischemic stroke: from data-driven approaches to personalized clinical decision-making

This thesis examines whether stroke care can move beyond the “average patient” toward decisions tailored to each individual. Stroke is the world’s second leading cause of death and a major cause of disability, yet treatment guidelines are based on average results across thousands of patients — results that may not apply to the specific person on the stretcher. The research developed and tested new tools, including an artificial intelligence system that estimates how much a particular patient is likely to benefit from clot-removal treatment, and a triage system that helped community hospitals deliver faster care. The studies show that personalising stroke treatment by combining clinical, imaging, and computational data improves outcomes, narrows hospital-level disparities, and supports more honest conversations with patients and families about both the benefits and the risks of treatment.