Adaptive wavelet discrimination of muscular noise in the ECG

The paper presents an adaptive time-frequency denoising algorithm. Main novelty is a running quasi- continuous scalo-temporal model of background activity built and subtracted from the ECG in order to yield a rectified representation of cardiac action. Our algorithm is based on the P, QRS and T wave borders automatically detected in the ECG and uses the information on expected local signal bandwidth to determine time-frequency regions containing cardiac representation. The complement is assumed to contain only the background activity representation and thus these values can be picked-up directly to the time-scale model of noise. The numerical tests performed with use of artificially noise-affected test signals reveal highly discriminative properties of the method. The amount of removed noise varies from 65% to 90% depending on input noise level.