A Quick Algorithm for QRS Noise Detection in ECG Based on Discrete Wavelet Transform

This paper presents an algorithm based on the discrete wavelet transform for noise detection from the ElectroCardioGraph (ECG) signal and recognition of abnormal heart beats. Wavelets provide simultaneous time and frequency information. The new algorithm detects noise of the R waves as well as noise of the Premature Ventricular Contraction (PVC) waves in the ECG signal. The wavelet transform decomposes the ECG signal into a set of frequency band. By using wavelet decomposition, we reduced the amount of data necessary to be processed by the algorithm to less than ten percent of the original data. The adaptive threshold algorithm is implemented with a value greater than that of R waves and less than the value of PVC. For the standard 24 hour Massachusetts Institute of Technology/Beth Israel Hospital (MIT-BIH) arrhythmia database, this algorithm correctly detects the noise 99.4 percent of the QRS complexes.