Automatic detection of conduction block based on time-frequency analysis of unipolar electrograms
F G Evans1, J M Rogers, W M Smith
1Department of Medicine, University of Alabama at Birmingham 35294, USA.
IEEE Transactions on Bio-Medical Engineering
|September 24, 1999
Summary
A new algorithm detects functional block during ventricular fibrillation (VF) using single electrode recordings. This method identifies reentrant substrates, crucial for understanding and treating lethal arrhythmias.
Area of Science:
- Cardiovascular Electrophysiology
- Computational Biology
Background:
- Lethal tachyarrhythmias like ventricular fibrillation (VF) are often sustained by functional reentry.
- Functional reentry occurs when an activation wave encounters a functional block, leading to wave rotation around excitable tissue.
Purpose of the Study:
- To develop and validate an automated algorithm for detecting functional block using single electrode recordings.
- To assess the algorithm's ability to identify regions of conduction block during VF.
Main Methods:
- Unipolar electrograms were recorded from a mapping array during electrically-induced VF in pigs.
- The Short-Time Fourier Transform (STFT) was employed to analyze electrograms and detect 'double potentials' indicative of functional block.
- Algorithm performance was evaluated against activation maps generated using a minimum conduction velocity criterion.
Main Results:
- The STFT algorithm demonstrated high accuracy in detecting conduction block.
- Sensitivity was 0.74 +/- 0.12, and specificity was 0.99 +/- 0.00.
Conclusions:
- An automated algorithm capable of detecting functional block from single electrogram recordings during VF has been developed.
- This algorithm offers potential applications in objective mapping data analysis and real-time localization of reentrant substrates.
