Conclusion

Two examples of the application of wavelet techniques to the representation and analysis of physiological waves have been presented in this article. The first example applies the Hermite wavelet basis associated with ECG data compression. The aim of the algorithm is to provide high-fidelity compression and reconstruction of the ECG waveform such that it does not introduce artifacts that could limit the clinical utility of the reconstructed waveform. To this end, a commercially available beat classification program was used to evaluate the performance of the compression program, rather than using the reconstruction rms error. Tested with the MIT-BIH Arrhythmia database, the classification error rate of the reconstructed waves is only 0.33% higher than that of the original data.

A wavelet-based algorithm has also been used to remove artifacts from NIBP oscillometric...