New Transform for Analyzing Atrial Electrograms
Author Information
Author(s): Ciaccio Edward J, Biviano Angelo B, Whang William, Coromilas James, Garan Hasan
Primary Institution: Columbia University
Hypothesis
Can a new data-driven transform improve the analysis of complex fractionated atrial electrograms (CFAE) compared to traditional Fourier analysis?
Conclusion
The new transform efficiently identifies CFAE features and is more robust to noise than Fourier analysis.
Supporting Evidence
- The ensemble basis is orthogonal and efficient for representation of CFAE components.
- The new transform outperformed Fourier analysis in detecting spectral components.
- The method showed a mean error of less than 10% in extracting synthetic drivers.
- Ensemble averaging was more accurate in identifying morphologic components.
- The transform is robust to phase noise and interference.
- Ensemble average spectra identified low-power transients better than Fourier spectra.
- Reconstruction errors were lower for ensemble averaging compared to Fourier.
Takeaway
This study created a new way to look at heart signals that helps doctors find problems better than the old method.
Methodology
The study used a data-driven basis and transform to analyze 216 recordings from 20 patients with atrial fibrillation.
Limitations
The study was limited to retrospective analysis and the frequency decompositions are only applicable to stationary signals.
Participant Demographics
10 patients with paroxysmal AF and 10 with persistent AF.
Statistical Information
P-Value
p ≤ 0.002
Statistical Significance
p ≤ 0.005
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website