A method is described for accurately measuring the volume and site of epilepsy surgery resections utilizing magnetic resonance imaging. Accuracy has been assessed using post-mortem studies, and both the intra- and interobserver variability is consistently less than 5%. The technique has so far been applied to 25 patients following a variety of operations for medically intractable epilepsy. It provides the crucially accurate baseline required for meaningful follow-up outcome studies of epilepsy surgery. Consequently, it should allow the development of more precise prognostic indices for such operations.