Clinical Diagnosis of Cardiac Disease Based on Support Vector Machine

Cardiac Diseases are very harmful to the human health. The application of electrocardiogram (ECG) is essential for the clinical diagnosis of cardiac diseases. The use of computers for accurately and quickly cardiac disease diagnosis has been a subject fervently pursued by both internal and external researchers. Therefore, it is great significant to explore even more accurate and higher-speed automatic ECG analysis method. Support Vector Machine (SVM) is a novel powerful machine learning method based on statistical learning theory, which is powerful for the characterization of small sample (nonlinearity, high dimension and local minima). In this paper, Clinical diagnosis of cardiac disease based on SVM is proposed. There are two input patterns of samples: 8-lead ECG in series and in parallel. All experiments are implemented on Pentium 350 MHz with 512 MB RAM. Matlab 6.5 is employ to solve the quadratic programming. Comparison in two input patterns based on SVM, the result shows that SVM method in parallel is highly reliable and accurate. It will have great potential application in clinical diagnosis.