Characteristics of magnetocardiography and electrocardiography in the time-frequency domain

As the carriers of electromagnetic information about the cardiac system, both magnetocardiography （MCG） and electrocardiography （ECG） are non-stationary signals （NSS）. The essential and crucial features of NSS are the local time-frequency characteristics （LTFCH）. Under the constraint of the uncertainty principle in the time-frequency domain （TFD）, the LTFCH of those non-stationary signals can be determined by evolutionary spectrum analysis （ESA）. In this paper, by analyzing the QRS complexes of both MCG and ECG data from healthy volunteers, we have derived some characteristic parameters （CHPs） in the TFD, such as the quality factor （Q） and center frequency （fz）. According to the experiment results, we believe that there are threshold ranges of Q andfz values in the dynamic behavior of the cardiac system for healthy projects. If the CHPs are in the proper ranges, the cardiac system is in a normal condition and homeostasis can be self-correcting. In contrast, if the values of the CHPs are less than the proper ranges, the homeostasis of the cardiac system is lost and some cardiac diseases may result. The assumption of the threshold ranges is verified by sample data from different cardiac diseases, and the results support the assumption. The CHPs are not only a kind of new reference criterion for ECG in clinical diagnosis, but also afford a technological route for the application of MCGs in cardiology.