摘要脑卒中临床表现多样，病因复杂，具有高发病率、高致残率、高复发率、高病死率和高经济负担五大特征。目前传统临床诊疗方法由于人力、时间等限制，存在患病和预后预测困难、诊断精确度低、治疗缓慢等问题。随着人工智能领域的深入研究和在医疗领域的应用，利用机器学习模型不仅能够较为准确地进行脑卒中预测和诊断，还可以识别危险因素，确定高危人群。本研究综述了机器学习算法的研究现状、脑卒中的危险因素识别和脑卒中预测常见的机器学习算法及在脑卒中风险预测中的研究现状和脑卒中风险预测的效果，为早期识别高危人群、采取有效的预防措施以及制定精确的治疗方案提供科学依据。

abstractsStroke has diverse clinical manifestations and complex causes, characterized by high incidence, high disability rate, high recurrence rate, high mortality rate, and high economic burden. Currently, conventional clinical diagnostic and treatment methods face challenges such as difficulty in predicting disease and prognosis, low diagnostic accuracy, and slow treatment due to limitations in manpower and time. With the in-depth research in artificial intelligence and its application in the medical field, machine learning models can not only predict and diagnose stroke more accurately but also identify risk factors and determine high-risk populations. This paper reviews the current research status of machine learning algorithms, the identification of stroke risk factors, common machine learning algorithms for stroke prediction, and the effectiveness of these algorithms in stroke risk prediction. Findings from this paper will help provide a scientific basis for the early identification of high-risk populations, the adoption of effective preventive measures, and the formulation of precise treatment plans.