Intelligent QRS typification using fuzzy clustering

For an automated ECG interpretation system, accurate QRS detection and typification including alignment, labelling, and dominant beat selection are essential for enhancing its performance. In this work we present a fuzzy clustering algorithm for intelligent beat labelling to decide whether each of the QRS complexes can be classified as the same cluster or not. Also for reliable classification, we have proposed efficiency feature sets that can best describe the morphology of the QRS complexes.