ELECTROCARDIOSIGNAL CLASSIFICATION MODEL TRAINING METHOD, ELECTROCARDIOSIGNAL CLASSIFYING METHOD AND APPARATUS

摘要

An electrocardiosignal classification model training method, an electrocardiosignal classifying method and apparatus, a computer-readable storage medium, and an electronic device. The electrocardiosignal classification model training method comprises: constructing a training sample set, each training sample containing an electrocardiosignal sample and a plurality of abnormality annotation tags (S210); acquiring a feature vector of the electrocardiosignal sample and abnormality tag vectors of the electrocardiosignal sample by means of a neural network of an electrocardiosignal classification model to be trained (S220); generating a tag correlation matrix according to the abnormality tag vectors, and generating an abnormality tag probability distribution vector according to the feature vector and the tag correlation matrix (S230); and according to the abnormality tag probability distribution vector and the plurality of abnormality annotation tags, training the electrocardiosignal classification model to be trained to obtain a trained electrocardiosignal classification model (S240). According to the present application, abnormal events of electrocardiosignals can be effectively classified by means of the model.