Research Of The Cardiac Pathology Recognition Technology Based On Learning

Cardiovascular disease is one of many human diseases that plague many people. Myocardial infarction is a terrible disease and large number of people dead due to this disease. The medical professions also need more research in the treatment of heart disease. Medicine and other disciplines, the cross-application practice can play a role. Research in computer science and medical diagnostic pattern recognition combining is already fruitful. It can help heart disease early prevention and real-time monitoring for heart health.Atrial hypertrophy is one of various heart diseases. But I did not find the relevant research results in pattern recognition field, due to the lack of data. It caused a huge obstacle for recognition research. This paper carried atrial hypertrophy specialized research and study the training and recognition algorithms in fewer cases of atrial hypertrophy. Compare the difference of classification accuracy of some different pattern classification methods in the case of small samples. This article main studied artificial neural networks and support vector machines (Support vector machine, SVM) used in the identification of atrial hypertrophy. Showing support vector machine (SVM) advances than other method in the case of small samples. I also improved SVM via classification integration. Improving the accuracy and credibility when integrate the SVM with Rejection.Due to the lack of ECG databases, it creates obstacles to artificial intelligence research and field of medicine. We always depend on foreign databases, like MIT-BIH database, MGH/MF database, etc. The ECG data used in this article is mainly from foreign authority ECG date base.