Acute hypotension episodes are one of the hemodynamic instabilities with high mortality rate that is frequent among many groups of patients. Prediction of acute hypotension episodes can help clinicians to diagnose the cause of this physiological disorder and select proper treatment based on this diagnosis. In this study Empirical Mode Decomposition of Mean Arterial Pressure (MAP) time series were calculated and some features such as statistical features of Intrinsic Mode Functions (IMFs) were extracted. Finally, a Support Vector Machine (SVM) was applied for classification of these features and prediction of acute hypotension episodes. The accuracy of prediction was 92% with Leave One Out cross validation method.