Background: This study was performed to elaborate an electrocardiographic (ECG) algorithm enabling assignment of an occluded coronary artery in acute myocardial infarction (AMI). Patients and Interventions: In 109 patients (age, 59±12 years) with AMI (pain onset, 3.6±1.7 h), coronary angiography with PTCA/stenting of the culprit lesion was performed. The diagnosis of AMI was confirmed by emergency coronary angiography and laboratory analyses. Admission ECG parameters (amplitude of R-wave, ST-segment deviation, presence of Q-wave, deflection of T-wave) in standard 12-lead ECG plus extended (V 3R to V 6R and V 7–9) leads were subjected to classification and regression tree (CART) analysis. Results: Continuous CART analysis assessed ST-segment deviations in V 2 and V 5R. AMI of the left anterior descending (LAD), right coronary artery (RCA) and left circumflex coronary artery (CX) were correctly classified in 94, 64, and 91% of cases, respectively. Dichotomised CART analysis assessed ST-segment deviations in V 2, V 5R, and aVF. True classification rates for LAD, RCA, and CX amounted to 84, 74, and 71%, respectively. Conclusions: Dichotomised CART analysis is a simple means of differentiation of CX from RCA occlusion during AMI.