Optimization-based reconstruction of depolarization of the heart

A new noninvasive method to obtain a priori data for solving the inverse problem of electrocardiography is suggested. The method employs a personalized 3D model of the patient built from MRI data and an electrophysiological model of the patient's heart (cellular automaton). The distribution of body surface potentials is calculated. The simulated ECG is "recorded" and compared with that measured for the patient. The parameter values of the cellular automaton are optimized in order to obtain the best possible correspondence between measured and simulated ECGs. The method provides a set of distributions of transmembrane voltages in the myocardium. These distributions can be used as a priori data for other types of inverse problems in electrocardiography. In this way valuable information about the cardiac electrical activity is obtained.