A heart monitor computes ST segment deviation as the difference in the value of an electrocardiogram signal at the ST and PQ points of a heartbeat. The ST point is found based on slope criteria and temporal criteria. The maximum (positive) slope of the QRS (maxQRS) is located. Within a preset window after the maxQRS point, the processor searches for a sample at which the second finite difference is less than a threshold that is a function of average QRS amplitude. If such a qualifying sample is found, the processor examines its location relative to the location of an adaptive window that is centered on a sample that is an adaptively determined distance from the peak of the R wave. If the qualifying point is within the adaptive window, it is chosen as the ST point. If the qualifying point is after the adaptive window, the ST point is set at the end of the adaptive window. Finally, if the qualifying point is before the adaptive window, the ST point is selected at the beginning of the adaptive window. The adaptive window is then updated according to an exponential average filter based upon the distance between the ST point and the R wave peak.