The technology

An Electrocardiogram (ECG) is an important diagnostic tool used to detect the activity of the heart. Parts of these signals can be obscured by baseline wander, which is noise cause by movement, breathing, or electrode impedance. It is important for this noise to be removed in order for clinicians to receive complete, uncorrupted information. Current methods involve using certain filters that are limited by frequency delineation or reference choice.

Independent Component Analysis (ICA) is used to make an estimation of the true baseline wander. This technology removes ECG baseline wander by combining adaptive notch filters with Blind Source Separation (BSS). This allows for the customization of ICA and a more complete removal of baseline wander than previous methods. Furthermore, the factors affecting the performance of the separation process are explored and improved in this invention. The following figure shows a signal comparison between the proposed method and a frequently used method, robust locally weighted regression. Significant error differences can be demonstrated. The invented methodology can be applied to remove noise from any other signal.