Objective: We wanted to develop a new method for processing dense spatially sampled unipolar electrograms, that yields physiological spatial distributions in activation-repolarization intervals.Background: Spatial heterogeneity in repolarization plays an important role in genesis and sustainment of cardiac arrhythmias. Reliable determination of repolarization times at high resolution on electrograms remains a difficult task. Methods: The new method describes the electrical activity on the heart's surface in terms of a physical electric field (E-field). Activation and repolarization times are calculated using an amplitude weighted average on QRS and T-waves. Activation-recovery intervals (ARI) obtained with this E-field method were validated against action potential durations (APD90) from monophasic action potential measurements. Relevance of the method was shown by calculating dense spatial distributions of beat-to-beat variability of repolarization (BVR) in myocardial infarction (MI).Results: E-field waveforms were more spatially homogeneous than the unipolar electrograms. The ARI strongly correlated with the APD90 (r2=0.94; p<0.001). Continuous activation and repolarization patterns were obtained, and the method removes non-physiological sharp spatial gradients obtained by currently adopted methods. Spatial BVR analysis revealed an increase in maximum BVR, spatial variance of BVR and spatial gradient of BVR in the border zone versus remote region of MI (all p<0.01). All BVR values were in pathophysiological range as described in other studies.Conclusion: The E-field method allows for a reliable determination of activation and repolarization times over the whole left ventricle. The new methodology confirmed that the border zone of myocardial infarction is a region with patches of increased variability of repolarization.