The ST-segment elevation myocardial infarction- cardiogenic shock (STEMI-CS) is one of the strongest factors in patient mortality within hospitals. This paper presents a hybrid machine learning based approach for predicting the risk of mortality in patients with STEMI-CS. The proposed method combines an efficient evolutionary differential search algorithm (DSA) with support vector machine (SVM) in risk prediction phase. The incentive mechanism of using DSA is to optimally tune the parameters of SVM to improve its prediction ability. With a test on a real-world benchmark dataset, the proposed DSA-SVM is confirmed to have significant improvement compared with multiple machine learning models.
