Cluster-Based Ensemble Learning Model for Rapid Detection of Aortic Dissection

Abstract

Background: Aortic dissection (AD) is a rare and high-risk cardiovascular disease with dangerous morbidity and high mortality, so it needs rapid diagnosis and timely treatment. However, due to its complex and changeable clinical manifestations and the lack of special symptoms and signs, it is easy to cause missed diagnosis and misdiagnosis.Methods: The data set used in this paper comes from 53213 patients, which collected from XiangYa Hospital in Hunan Province from 2008 to 2016. The data includes 802 patients with aortic dissection and 52411 patients with non-aortic dissection. In order to help clinicians predict AD, we designed an ensemble learning model based on clustering: Cluster Random Under-sampling Smote-Tomek-link Bagging (CRST-Bagging). This model combines the advantages of clustering-based compound resampling (CRST) method and Bagging ensemble classifier. It achieves good results on aortic dissection data sets.Results: The model validates the effectiveness of the CRST sampling method on the AD data set. We compared the CRST-Bagging model with the classical ensemble models RUSBoost and SMOTE-Bagging on the AD data set. The experimental results show that the CRST-Bagging model has the best performance in the detection of AD. Model’s accuracy and recall rate are 83.6% and 80.7% respectively. And the F1 value is 82.1%, which is 4.8% and 1.6% higher than that of RUSBoost and SMOTE-Bagging model.Conclusions: The model proposed in this paper can be used as an auxiliary diagnostic model of AD to provide reference for clinical medicine. The model can also help doctors to judge whether patients need further imaging examination. Thus help to reduce the rate of clinical misdiagnosis and missed diagnosis of AD.