Modeling of Cardiac Activation and Surface Potential Distribution: Correlation with Computer Analysis of St-T Segment.

A method for determining the potentials due to internal sources within an irregularly shaped conducting medium as applied to electrocardiography is presented in the thesis. Due to the complexities of the analytical expressions relating the internal source, i.e., the heart, to the surface potentials, these equations cannot be solved analytically to obtain the exact solutions. Thus the equations are expressed by discrete sum approximations and a numerical method is used to solve the surface potential distributions. A complete treatment of the time dependency of the problem in dielectric material is presented and it is shown that the problem can be treated as a quasi-static one. From the computed surface potential distribution, the st and ard 12 lead electro-cardiogram is synthesized by using realistic action potential models and ventricular activation sequence to simulate the depolarization (QRS complex) and repolarization (T wave) of the ventricles. The ischemia is simulated by using ischemic action potentials in place of the normal action potentials, and varying the size and location of the ischemic region of the ventricles. The results are studied to correlate the effect of the ischemia on the surface recorded ECG for better underst and ing of the morphologic changes of the ST-T segment. New computer algorithms for detecting the QRS complex and segmenting the T wave were developed for operator independent ST-T segment analysis. Based on a new template matching technique, the ST-T analysis program detects abnormal beats with ST-T changes and reports an extensive set of parameters.Ph.D.Biomedical engineeringUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/161123/1/8621321.pd