The effects of different corrections for background activity in the estimation of low organ uptake of radiopharmaceuticals have been examined using myocardial perfusion imaging agents. Estimates of myocardial uptake of 99mTc-labelled MIBI and tetrofosmin were made both at rest and after exercise. Patients were given one or other of the agents (12 MIBI; 17 tetrofosmin) and the measurements at rest and after exercise were made within a week of each other using a planar gamma camera method incorporating an attenuation-corrected, geometric mean technique. Myocardial uptakes were estimated using two different background corrections and also with no background subtraction. Mean values were in the range 1.3 to 3.0% and showed that, for both agents, uptakes estimated with and without background correction could differ by a factor of two. Although the study was not designed to compare myocardial uptakes of the two agents, a background correction which accounted separately for activity in tissue over- and under-lying the heart resulted in similar mean values for tetrofosmin (1.7% both at rest and after exercise) and for MIBI (1.8% rest; 1.9% exercise). For both agents, no significant difference was observed between myocardial uptakes at rest and after exercise measured at about two hours post-injection.