A basic study on algorithm for automatic diagnosis by magnetocardiography

Final goal of this study is to develop an automatic diagnosis algorithm of the cardiac diseases using magnetocardiography (MCG) measured by SQUID system. To estimate spatial distribution and changes in time of cardiac activities, we have developed a spatial frequency analysis method for MCG. By simulations with a simple cubic model for ventricle, it was found that the spatial phase spectrum of MCG indicates the movement and the direction of the cardiac excited wave front, and the amplitude spectrum indicates the spread of the excited wave front. We preliminarily applied this method to the MCG of a healthy subject, before examining the validity of this method to the abnormal cardiac excitation. The traces of the excited wave front that are considered by the analyzed result were not inconsistent with the present electrophysiological knowledge. It is suggested that this presented method can trace the spread of cardiac excitation with sufficient accuracy.