An atrial fibrillation detection algorithm combining Quiet Interval Analysis and Baseline Crossing Analysis was designed to achieve high specificity for use in implantable atrial defibrillators.
The purpose of study was to evaluate the performance of an atrial fibrillation (AF) defection method for use in an implantable atrial defibrillator. Since AF is not immediately life threatening, a detection algorithm can be designed to be highly specific to avoid inappropriate shock delivery. Wideband electrograms (EGMS) were recorded from an acute bipolar ventricular lead (RV), and from a vector formed by two multipolar mapping leads located in the right atrium (RA), and coronary sinus (CS) in humans. A filtered RV EGM was used to provide ventricular sensing for R-wave detection. A filtered RA-CS EGM was used for AF detection. The gain of each sensing channel was automatically adjusted. The AF detection algorithm consisted of two independent serial tests, Quiet Interval Analysis (QIA) and Baseline Crossing Analysis (BCA). To obtain high specificity, if was required that both algorithms detect the analyzed rhythm as atrial fibrillation to classify if as atrial fibrillation.
Kim et al. (Tue,) studied this question.