Automatic Identification of Cardiac Activations in the Multi-Channel Intracardiac ECGs

A novel algorithm is presented in this paper, which can precisely identify the cardiac activations (e.g., A, V, H waves) in the multi-channel intracardiac electrogram (IECG) signals. The aim of this work is to achieve an automatic analysis during the electrophysiological (EP) study. It saves an amount of the time for the physicians compared with manual work, especially for the long-duration diagnosis or operation. The development of the algorithm complies with the various stimulation protocols in the EP lab. The effective digital signal processing techniques and the proper application of the relationship between the surface ECG (SECG) and intracardiac ECG enable the algorithm to achieve the satisfied sensitivity. Finally, 11 patients´ data including more than 6000 meaningful cardiac activations are randomly collected from an EP (electrophysiology) lab to evaluate the algorithm. A comparison is made between the manual annotation of the physicians and the result of the algorithm. The sensitivity of the V, A, H wave detection reaches 98.99%, 96.64%, and 97.31% respectively.