A machine learning approach for predictings stroke  

作　　者：Yubo Fu 

机构地区：[1]Hasan School of Business,Colorado State University Pueblo,Pueblo,CO 81001,USA

出　　处：《Medical Data Mining》2024年第3期8-16,共9页TMR医学数据挖掘

摘　　要：Background:Stroke is one of the most dangerous and life-threatening disease as it can cause lasting brain damage,long-term disability,or even death.The early detection of warning signs of a stroke can help save the life of a patient.In this paper,we adopted machine learning approaches to predict strokes and identify the three most important factors that are associated with strokes.Methods:This study used an open-access stroke prediction dataset.We developed 11 machine learning models and compare the results to those found in prior studies.Results:The accuracy,recall and area under the curve for the random forest model in our study is significantly higher than those of other studies.Machine learning models,particularly the random forest algorithm,can accurately predict the risk of stroke and support medical decision making.Conclusion:Our findings can be applied to design clinical prediction systems at the point of care.

关 键 词：medical decision making machine learning predictive modeling STROKE imbalanced data 

分 类 号：R74[医药卫生—神经病学与精神病学]