Electrocardiogram (ECG) is the procedural electrical activity recording of the heart that arises from the heart muscle's electrophysiological pattern. But in clinical atmosphere during acquisition, the ECG signal is corrupted with various types of artifacts. The parting of the preferred signal from noises caused by artifacts such as muscle artifacts, power-line interference, base-line wandering and motion artifacts is a big covenant. Among these noises a power-line interference of 50 Hz frequency is more severe and fluctuate the signals morphological appearances. There are various tools like wavelet transform and empirical mode decomposition (EMD) being used for filtering other than conventional filters. EMD based noise cancellation is a fully signal dependent approach and adaptive in nature that can be used for real-time applications. This paper deliberates the comparative analysis of EMD based filtering methods for noise cancellation in ECG signal under 48–51 Hz of frequency under varying noise amplitudes. © 2017