We hypothesized that not only the location but also extent of myocardial ischemia (area at risk, AAR) during acute myocardial infarction could be estimated from the spatial distribution of ST-segment deviations in the ECG. Standard 12-lead ECGs and SPECT images were obtained from 75 patients. Precordial ST-segment deviations were organized according to spatial orientation. An ST-profile was obtained: a semicircle of ST-deviations centered on the ST-injury vector. Spatial features of the profile were calculated, the R wave vector magnitude was obtained, and an automated method of patient separation into culprit artery groups was developed. A linear regression model between optimal features and AAR yielded: r = 0.64(p