Acute myocardial infarction positioning method based on GA-ELM (genetic algorithm-extreme learning machine) mixed model

Abstract

The invention relates to an acute myocardial infarction positioning method based on a GA-ELM (genetic algorithm-extreme learning machine) mixed model. The method comprises the following steps of (1) signal preprocessing: removing baseline drift by a median filtering algorithm; removing power frequency interference by a band rejection filter; removing electromyographical interference through low-pass filtering; removing electromyographical interference signals by a Chebyshev digital low-pass filter; finally, obtaining high-quality electrocardiosignals; (2) heart beat segmentation: using DB6 wavelets as mother wavelets; positioning R wave peaks by a detection algorithm based on wavelet transform; using positioned R wave peak value points as reference points; respectively forwards and backwards selecting original sampling data being 250 ms and 400 ms as feature vectors; (3) model training: optimizing parameters randomly selected from an ELM by a GA; then, training an ELM network by the obtained optimal parameter; finally, positioning myocardial infarction by the trained ELM network. The method has the advantages that the setting of a plurality of network training parameters is not needed; the structure is simple; the training speed is high.

Description

technical field [0001] The invention belongs to the technical field of heartbeat detection and classification, and in particular relates to a method for locating acute myocardial infarction based on a GA-ELM hybrid model. Background technique [0002] Cardiovascular disease is one of the diseases that seriously threaten human health. With the development of social economy and the change of national lifestyle, as well as the aging of the population and the acceleration of urbanization, the number of cardiovascular diseases in my country has continued to increase rapidly. acute myocardial infarction [1] Refers to the lesions of the coronary arteries that nourish the myocardium, the blood supply of the coronary arteries is sharply reduced or completely interrupted, and the myocardial cells are damaged or even necrotic caused by severe and persistent ischemia and hypoxia of the corresponding myocardium. Myocardial infarction has extremely high mortality and disability Rate. Fo...

AI Technical Summary

Problems solved by technology

There are many symptoms of myocardial infarction, but there are also many patients who only have mild symptoms or no symptoms at all, and it is easy to miss the best time for treatment due to misdiagnosis. important

[0003] Researchers at home and abroad have proposed various automatic detection and location algorithms for acute myocardial infarction, such as the random forest hierarchical classification method, which realizes the detection of myocardial infarction. In order to apply to the real-time analysis system, the researchers increase the number of features layer by layer to Reduce the time required for classification; use convolutional neural network (Convolutional Neural Networks, CNN) method to realize the detection of myocardial infarction; use feed-forward neural network (Feed-Forward Neural Network, FFNN) and SVM to realize the detection of myocardial infarction, However, the above-mentioned detection method of myocardial infarction does not carry out the location analysis of the myocardial infarction site. The ECG records obtained under different leads have irreplaceable clinical value. Artificially reducing the lead data will lead to the loss of effective information and directly affect the location of myocardial infarction. the accuracy of

Images