Removal of ECG contamination from diaphragmatic EMG by nonlinear filtering

We describe a method to suppress electrocardiogram (ECG) contaminations in the diaphragm electromyographic (EMG) signals obtained through intraoesophageal electrodes. A nonlinear filter technique originally developed for noise reduction in deterministically chaotic signals is used. The performance of the method was investigated using computer simulations. Multitaper spectral analysis was performed to assess quantitatively the difference between the simulated pure EMG signal and the processed EMG signal. The method is successfully applied to recordings from healthy subjects during deep inspiration.