An integration of improved median and morphological filtering techniques for electrocardiogram signal processing

Electrocardiogram contains a wealth of diagnostic information normally used to guide clinical decision making for proper diagnosis of cardiovascular diseases disorders. ECG is often contaminated by noises and artifacts that can be within the frequency band of interest and can manifest with similar morphologies as the ECG signal itself. Baseline correction and noise suppression are the two important pre-requisites for conditioning of the ECG signal. This paper presents a novel morphological filtering technique for removing baseline drift using non-flat structuring element. Further, to achieve noise suppression an improved median filtering technique is applied using mask of variable sizes. Depending upon the degree of impulse noise contamination the mask size may vary to a maximum of 1×11. The residual noise left after this stage is then filtered out using morphological filtering. Simulation results show noteworthy improvement in baseline correction and noise filtering in comparison to other proposed morphological filtering based approaches in the literature.