We compare three inverse algorithms - zero-order Tikhonov, first-order Tikhonov, and first-order Generalized Eigensystem regularization - for their ability to estimate the endocardial potentials from measured probe potentials and known geometric data. For each method, the Composite Residual Error and Smoothing Operator (CRESO) technique was used to estimate the regularization parameter. For the preliminary data examined here, higher-order regularization produced higher correlation coefficients between the estimated and measured endocardial electrograms.
- GeneraUzed Eigensystem method
- Inverse problems
- Tikhonov regularization
ASJC Scopus subject areas