Prediction of Acute Hypotension Episodes in Patients Taking Pressor Medication Using Modeling of Arterial Blood Pressure Waveforms

摘要：

This paper presents an effective technique for solving the prediction problem presented in Event-1 of Physionet Challenge 2009. In this challenge, the prediction of occurrence of an Acute Hypotension Episode (AHE) is to be made (in patients receiving pressor medication) using clinical data and medical signals prior to the start of a forecast window. The technique proposed in this paper uses time domain features along with principal components of the Arterial Blood Pressure (ABP) waveform averaged over beats lying in each non-overlapping 60s interval for 1.5 hours prior to the start of the forecast window. Classification is performed with a simple Linear Support Vector Classifier (LSVC) after feature selection through genetic algorithms. This method uses only 5 features to give a perfect score of 10/10 over data in event-1 test set.

作者: Fayyaz A.Afsar

作者单位: Department of Computer and Information Sciences Pakistan Institute of Engineering and Applied Sciences (PIEAS), PO Nilore, Islamabad, Pakistan

会议类型: 国际会议

会议名称: The 4th International Conference on Bioinformatics and Biomedical Engineering(第四届IEEE生物信息与生物医学工程国际会议 iCBBE 2010)

会议地点: 成都

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-18（万方平台首次上网日期，不代表论文的发表时间）