For a successful diagnosis, electrocardiogram (ECG) must be cleaned from the power line interference (PLI) and the baseline wander (BW). In this framework, an adaptive rate linear phase filtering method is suggested. The system parameters adapt automatically by monitoring the incoming signal variations. It adjusts the acquisition rate and the noise-suppressing filter order and renders the filtering of lesser samples by filters of inferior taps compared to the conventional counterparts. The applicability is demonstrated by using the MIT-BIH ECG database. The accuracy of the system is measured in terms of the signal to noise ratio (SNR). Results show that while providing a comparable output quality, the proposed method achieves a two times compression gain and notable computational efficiency over the conventional counterparts. A superior precision of the suggested solution over the state-of-the-art counter equals is also demonstrated.