Electrocardiograph (ECG) Recognition Based on Graphical Fusion with Geometric Algebra

Since the emergence of extensive multivariate data, feature fusion has been more and more important. Most conventional methods of feature fusion do not pay sufficient attention to inherent geometric properties of data, even in the case where the data have spatial features. Therefore, how to fuse multiple features in a more intuitive way is still an open problem. A new strategy of parallel graphical feature fusion is introduced in this paper. The original time series was first divided into subspace, then extracting multi-dimensional features in each subsequence space and mapping to points in high-dimensional space, and giving the geometry object representation in high dimensional feature space of the feature points, finally extracting the geometric features of the geometry object. One task of the research is to use geometry algebraic language to describe geometric objects of feature space and using geometry calculation to extract geometric features. In the two-dimensional space and three-dimensional space to build multi-ECG morphology triangles, determining the most effective parameters of the geometric structure and the pattern classification were done on five types of ECG time series of the MIT/BIH arrhythmia database. It got a high accuracy.