Abstract
The equivalent-dipole method, in which one or more electric dipoles are determined so that they best approximate electrical activities of the heart is one of the representative methods of solving the inverse problem in electrocardiography. Resultant equivalent dipoles are affected by several factors, such as simplification of the human torso, noise in the measured potentials, fluctuations in the reference potential and displacement of the electrode locations.
To investigate the effects of these factors, we performed computer simulations and have come to the following conclusions. (1) Considering the calculation time as well as the precision of the results, the infinite-medium model, in which the body is immersed in a conductor with the same conductivity as that of the inside of the torso, is most suitable. (2) When the potential distributions are measured by 128 electrodes arranged on the chest and the back, dipole estimation error is less than 5mm provided that noise and offset voltage error be 20% of the RMS value of the body surface potential distribution. (3) When the excitation front is located near the body surface, dipole estimation error is comparable to the electrode location error.
