Evaluation of Endocardial Electrograms Fractionation Complexity in Human Using Statistical Pattern Recognition

摘要

Identification of complex fractionated atrial electrograms (CFAEs) sites is crucial for the development of new AF ablation strategies. CFAE may represent the electrophysio-logical substrate for atrial fibrillation (AF). Progress in signal processing algorithms is the key part of this task. Algorithm for automated description of atrial electrograms (A-EGMs) fractionation based on wavelet transform and several feature extraction methods and statistical pattern recognition was proposed and methodology of A-EGM processing was designed and tested. The algorithms for signal processing, description and classification were developed and validated using a representative set of 1.5 s A-EGMs (n = 113) ranked by 3 independent experts into 4 classes of fractionation: 1 - organized atrial activity; 2 - mild; 3 - intermediate; 4 - high degree of fractionation. New feature extraction and classification algorithms used and tested here showed mean classification error over all classes ~ 8.3%, and classification error of highly fractionated A-EGMs of ~ 9%. Such fully automatic algorithms for A-EGMs complexity description that do not need operator interaction may be easily incorporated into mapping systems to assist and guide AF substrate ablation.