Title :
Minimum distance to origin: a method for choosing the number of expansion modes for the generalized eigensystem method
Author :
Throne, RD ; Olson, LG ; Hrabik, TJ
Author_Institution :
Dept. of Electr. Eng., Nebraska Univ., Lincoln, NE, USA
Abstract :
The authors have previously presented the generalized eigen-system (GES) approach as an alternative to truncated singular value decomposition and zero, first, and second order Tikhonov regularization methods for the ill-conditioned inverse problem of electrocardiography. The results from the GES methods were generally superior when the optimal parameters were used for each method. In this study the authors introduce the Minimum Distance to Origin (MDO) method for choosing the number of GES expansion vectors
Keywords :
electrocardiography; inverse problems; physiological models; singular value decomposition; vectors; ECG model; electrocardiography inverse problem; electrodiagnostics; expansion vectors number; generalized eigen-system approach; ill-conditioned inverse problem; minimum distance to origin method; optimal parameters; second order Tikhonov regularization methods; truncated singular value decomposition; Cardiography; Electrodes; Finite element methods; Heart; Inverse problems; Measurement standards; Mechanical engineering; Singular value decomposition; Solid modeling; Torso;
Conference_Titel :
Computers in Cardiology 1998
Conference_Location :
Cleveland, OH
Print_ISBN :
0-7803-5200-9
DOI :
10.1109/CIC.1998.731969
