Acute ischemic stroke (AIS) is characterized by high rates of disability, mortality, and recurrence, posing a significant burden on patients and society. In the era of big data, predictive models are increasingly used in patient diagnosis, treatment decisions, prognosis management, and healthcare resource allocation, highlighting their growing importance. Machine learning methods have become a crucial tool for predicting the prognosis of AIS patients and have been widely applied. This review explores recent advancements in the study of AIS prognosis prediction, focusing on machine learning methods. It discusses current issues and challenges faced by machine learning models, aiming to provide new insights and references for methods of early assessment and prediction of prognosis outcomes in AIS patients.

急性缺血性脑卒中（AIS）具有高致残率、高病死率及高复发率等特点，给患者及社会造成沉重的负担。随着大数据时代的到来，预测模型在患者的诊治决策、预后管理以及卫生资源配置等方面的应用越来越多，其价值也愈发重要。机器学习方法是预测AIS患者预后的重要方法之一，且已广泛应用。本文以机器学习方法为重点，就AIS预后预测研究的最新进展予以综述，并提出机器学习预测模型目前所面临的问题与挑战，为AIS患者预后结局早期评估与预测在方法上提供新的思路和参考。