A modified generalization eigensystem technique for the inverse problem of electrocardiography

The authors have previously presented the generalized eigensystem (GES) approach as an alternative to truncated singular value decomposition and Tikhonov regularization methods for the ill-conditioned inverse problem of electrocardiography. GES worked well on the eccentric sphere model problem with additive body noise. Here the authors investigate the effects of errors in geometry using GES, a modified generalized eigensystem method (tGES), truncated singular value decomposition, and zero order Tikhonov regularization. tGES, which allows the expansion coefficients to decay in a manner similar to Tikhonov regularization, shows the best performance overall