Method and system for identifying atrial fibrillation electrocardiosignals through weighted multi-scale limited penetrable visibility graph

The invention belongs to the field of atrial fibrillation electrocardiosignal processing, and provides a method and system for identifying atrial fibrillation electrocardiosignals through a weighted multi-scale limited penetrable visibility graph. The method comprises the following steps: acquiring electrocardiosignals with a set length; extracting original features of the electrocardiosignals to obtain corresponding original feature vectors; converting the electrocardiosignals into a network form, and extracting corresponding network features so as to obtain corresponding network feature vectors, wherein the network features comprise newly-added local efficiency entropies; and fusing the original feature vectors and the network feature vectors and then inputting the fused vectors into a machine learning model so as to obtain a category identification result of the electrocardiosignals, wherein the categories of the electrocardiosignals comprise atrial fibrillation electrocardiosignals, ...