The USC Andrew and Erna Viterbi School of Engineering USC Signal and Image Processing Institute USC Ming Hsieh Department of Electrical and Computer Engineering University of Southern California

Technical Report USC-SIPI-191

“Forward and Inverse Problem Studies For Electrocardiography Based on a Reconstructed Realistic Canine Model From CT-scan Sections”

by Shan Gao and Chrysostomos L. Nikias

October 1991

A realistic canine anatomy model with high resolution has been reconstructed from CT-scan sections by means of digital image signal processing technique. Based on this model, the forward transfer coefficient matrix ZMH has been computed using the surface integral method, and the inverse solutions in the presence of measurement noise on torso surface from the SVD method have been tested. For the consideration of clinical utility, the internal organs of canine were included and a set of real canine ECG signals were used in this study.

Results from the forward problem using surface integral method and the inverse problem using regularization, SVD and recursive methods based on a homogeneous sphere model were also presented in this thesis.

To download the report in PDF format click here: USC-SIPI-191.pdf (4.7Mb)