The effect of 50/60 Hz notch filter application on human and rat ECG recordings.

Power-line interference is always present in indoor biopotential measurements, even when its extremely low magnitude makes it imperceptible. In special situations, this kind of interference can be neglected, but this is not a general rule. In laboratory experiments and clinical analysis, it is hard (and expensive) to isolate the subject of measurement from electrical fields produced by a power line. In human biopotential recordings, it is common practice to apply a 50/60 Hz notch filter to reduce this kind of interference. In such cases, there is no considerable distortion observed on the recorded signal. However, experiments showed that it is not true for rat ECG recordings. Several kinds of notch filters (analog and digital) were implemented for evaluation of the distortion caused on ECG signals. These filters were applied to ECGs of humans and rats and then distortion estimates were computed from their resulting signals. The comparison of these estimates showed that, as experimentally observed, rat ECG signals are significantly distorted and deformed when a 50/60 Hz notch filter is applied to them, while human ECGs are not. The major goal of this paper is to show that the use of a notch filter for power-line interference rejection, when applied to rat ECG recordings, can severely deform the QRS complex of such signals, warning researchers against its non-deliberate usage.