An estimation of defibrillation threshold based on electrogram signal disorganization.

摘要

A major focus of Implantable Cardioverter Defibrillator (ICD) research has been to reduce defibrillation shock energy to prolong battery life. We investigated whether the degree of disorganization of electrograms are correlated with defibrillation shock outcome. The study data were recorded during device implantation. A total 111 data segment from 41 patients were analyzed.; Haar wavelet-based QRS beat detection was developed to detect sinus and ventricular fibrillation (VF) beats. This method identified beats with an accuracy of 99.35%. A novel slope reversal algorithm identified QRS waves.; A new metric, Entropy Index (EI), based on Approximate Entropy was developed to quantify the disorganization of the electrograms. It discriminated successful from failure episodes with a sensitivity of 81% and a specificity of 83%.; Another metric, Similarity Index (SI), based on a cross correlation was also examined. Following the trend set by EI, SI quantified disorganization and predicted shock outcome. It classified successful episodes from failure with a sensitivity of 60% and a specificity of 74%. Using the indices, a relationship between defibrillation shock energy and VF complexity was determined.; The study concluded that the organization of VF is indeed a potential tool to predict the minimum energy required for successful defibrillation.