DocumentCode
2127698
Title
A comparison of zero and first order Tikhonov regularization for an inhomogeneous volume
Author
Lowther, DA ; Throne, RD ; Olson, LG ; Windle, Jr.
Author_Institution
Dept. of Electr. Eng., Nebraska Univ., Lincoln, NE, USA
fYear
2001
fDate
2001
Firstpage
541
Lastpage
544
Abstract
In this study, we examine the effects of tissue inhomogeneities on epicardial estimates when using zero-order Tikhonov and first-order Tikhonov regularization. Epicardial potentials measured during an in-vivo swine experiment were projected to 96 body surface locations using an inhomogeneous transfer matrix constructed using a finite-element model of the swine made from CT scans. It was assumed that the lungs, fat and remaining volume had conductivities of 50, 40 and 220 μmho/mm. The inverse estimates were then computed from the body surface potentials where noise had been added to simulate measurement errors. The inverses were computed using the same inhomogeneous transfer matrix used for the the forward projection, and with a homogeneous transfer matrix
Keywords
bioelectric potentials; biological tissues; biology computing; computerised tomography; electrical conductivity; electrocardiography; finite element analysis; inverse problems; lung; measurement errors; transfer function matrices; CT scans; body surface potentials; computerized tomography; electrical conductivity; epicardial potentials; fat; finite element model; first-order Tikhonov regularization; forward projection; homogeneous transfer matrix; inhomogeneous transfer matrix; inhomogeneous volume; inverse estimates; lungs; measurement errors; noise; swine; tissue inhomogeneities; zero-order Tikhonov regularization; Amplitude estimation; Computational modeling; Computed tomography; Conductivity; Electrodes; Finite element methods; Heart; Laplace equations; Lungs; Torso;
fLanguage
English
Publisher
ieee
Conference_Titel
Computers in Cardiology 2001
Conference_Location
Rotterdam
ISSN
0276-6547
Print_ISBN
0-7803-7266-2
Type
conf
DOI
10.1109/CIC.2001.977712
Filename
977712
Link To Document