Electrocardiogram classification method and system based on dynamic learning

The invention provides an electrocardiogram classification method and system based on dynamic learning, and relates to the technical field of electrocardiogram classification, and the method comprises the steps: obtaining a plurality of 12-lead electrocardiogram data and corresponding category labels; preprocessing the 12-lead electrocardiogram data, extracting a plurality of electrocardio characteristics of the electrocardiosignal from the preprocessed 12-lead electrocardiogram data, and selecting a preset number of electrocardio characteristics to construct a characteristic subset according to mutual information between each electrocardio characteristic and a category label; building a random forest according to the feature subsets and the category labels, and converting the random forest into a single decision tree to form a final cardiovascular disease detection model; and classifying the received to-be-detected electrocardiogram data by using the constructed cardiovascular disease detection model to obta