A new method for atrial electrical activity analysis from surface ECG signals using an energy ratio measure

A new method called separation using maximum energy ratio (SUMER) is introduced. Using 12 lead ECG signals, SUMER tries to separate the atrial activity from the ventricular activity. Relying on the assumption that the atrial activity can be reconstructed from a linear combination of 12 lead ECG signals, SUMER looks for the combination that will give us the best representation of the atrial activity. A cost function that is the energy ratio between different segments in the ECG signal is created. Forcing the linear combination to find the maximum possible cost function gives us the desired combination. The method presents better separation performances than the ICA method, and provides tools for the next phase of P wave detection.