An integrated method for clustering of QRS complexes is presented which inc
ludes basis function representation and self-organizing neural networks (NN
's), Each QRS complex is decomposed into Hermite basis functions and the re
sulting coefficients and width parameter are used to represent the complex.
By means of this representation, unsupervised self-organizing NN's are emp
loyed to cluster the data into 25 groups. Using the MIT-BIH arrhythmia data
base, the resulting clusters are found to exhibit a very low degree of misc
lassification (1.5%). The integrated method outperforms, on the MIT-BIH dat
abase, both a published supervised learning method as well as a conventiona
l template cross-correlation clustering method.