Related Experiment Videos
Bandwidth-induced errors in parameters used for automated activation time determination during computerized
C F Pieper1, G Lawrie, R Roberts
1Department of Medicine, Baylor College of Medicine, Methodist Hospital, Houston, TX 77030.
Pacing and Clinical Electrophysiology : PACE
|February 1, 1991
Summary
The study found that bipolar electrograms (BP) require less bandwidth than monopolar electrograms (MP) for accurate cardiac mapping. Preserving 95% signal power in BP is sufficient, while MP needs 99.9% for comparable accuracy in determining local activation times.
Area of Science:
- Biomedical Engineering
- Cardiac Electrophysiology
- Signal Processing
Background:
- Accurate determination of local activation time is crucial for intraoperative cardiac mapping.
- Bipolar electrograms (BP) and monopolar electrograms (MP) utilize different parameters (peak time and largest negative slope, respectively) for activation time determination.
- The impact of signal bandwidth on the accuracy of these parameters is not fully understood.
Purpose of the Study:
- To investigate the dependence of BP peak time and MP largest negative slope (LNS) determination on signal bandwidth.
- To quantify the errors introduced by filtering electrogram signals at different bandwidths.
- To establish optimal bandwidth requirements for accurate intraoperative cardiac mapping using BP and MP.
Main Methods:
- Compiled a database of fast MP and BP from intraoperative epicardial sock array recordings in humans.
- Determined signal bandwidths based on retaining 95%, 99%, and 99.9% of total signal power.
- Utilized Fourier transform to filter signals and computed errors (mean square error, MSE) in parameter determination.
- Characterized errors with respect to percent power bandwidth and frequency content.
Main Results:
- Filtering BP at 200 Hz caused a 1-ms peak shift in 17% of electrograms, with minimal shifts beyond that.
- Comparable shifts in MP LNS occurred when filtering at 400 Hz.
- Retaining 95% of BP power rarely shifted the peak time by more than 1 ms (0.9%).
- For MP, retaining 99.9% of power was necessary for similar accuracy (3.6% shift > 1 ms).
Conclusions:
- Bipolar electrograms possess greater high-frequency signal power than monopolar electrograms.
- Accurate determination of the LNS in MP requires a wider bandwidth compared to determining the peak in BP.
- These findings have implications for optimizing filtering strategies in computerized intraoperative cardiac mapping to ensure accurate local activation time measurement.