机器学习在脑卒中预测中的研究进展

Research Advances in Machine Learning for Stroke Prediction

机器学习在医疗领域展示了无法比拟的优越性,其可以提供快速准确的预测结果。利用机器学习方法对脑卒中进行预测,有助于临床医务人员做出有效的临床决策,降低脑卒中的发病率和病死率。本文对机器学习在脑卒中领域的研究进行综述,重点关注其对脑卒中复发、预后以及恢复预测。支持向量机、极端梯度提升、深度神经网络、K近邻等显示了较好的预测性能,值得推广应用。

Machine Learning has demonstrated unparalleled superiority in the medical field by providing fast and accurate prediction results.Prediction of stroke using machine learning methods can provide clinical decision makers with effective clinical decisions to reduce the morbidity rates and case-fatality rates.This paper focuses on the research of machine learning in the field of stroke,focusing on stroke recurrence,prognosis and recovery prediction,and finds that support vector machine,extreme gradient improvement,deep neural networks,K-nearest neighbor,etc,show good prediction performance and are worth promoting for application.