Key result
Using a Wiener filter as post-processing for Independent Component Analysis improved the accuracy of fetal R wave peak detection, reducing the relative error rate to 14% compared to ICA alone.
Why the study?
Does using a Wiener filter for post-processing in ICA improve noise rejection and R wave peak detection accuracy in fetal magnetocardiograms?
Does using a Wiener filter for post-processing in ICA improve noise rejection and R wave peak detection accuracy in fetal magnetocardiograms?
Absolute Event Rate: 14% vs 100%
Using a Wiener filter as post-processing for Independent Component Analysis significantly improves the accuracy of R-wave peak detection in fetal magnetocardiograms by effectively reducing noise.
May improve fetal R-wave detection in magnetocardiography; leaves open prospective validation before clinical adoption.
We measured fetal magnetocardiograms (fMCGs) to analyze variability in the heart rate of a fetus. The heart rate variability was calculated from the interval of the R wave peak of fMCGs. However, the fMCGs had a very weak signal that was buried in noise, and noise rejection was necessary. Independent component analysis (ICA) is a useful method of distinguishing signal and noise components. However, it is difficult to separate noise from the signal, when the measured data have a low signal-to-noise ratio. We propose a method that uses a Wiener filter for post processing in ICA. The Wiener filter can reduce the amount of noise present in a signal by comparison with estimates of the required noiseless signal. The required signal in this study was created by measuring the data. The proposed method was evaluated for 15 fMCGs. As a result, the accuracy of R wave peak detection was improved by using the Wiener filter for post processing in ICA.
No takes yet. Share an insight, caveat, or question.
Sato et al. (2011) studied Healthy pregnancy (fetal magnetocardiogram) (n=15). Wiener filter as post-processing in Independent Component Analysis (ICA) vs. Independent Component Analysis (ICA) alone was evaluated on Error rate in detection of R wave peak (relative to ICA alone). Using a Wiener filter as post-processing for Independent Component Analysis improved the accuracy of fetal R wave peak detection, reducing the relative error rate to 14% compared to ICA alone.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: