Because commercially available camera-based methods are not optimized, they fail to account for dose infiltration, table attenuation and correspondence between time of injection and starting the camera. We have developed a more optimized technique to calculate camera-based clearances and applied this technique in the design of a camera-based clearance method for 99mTc-MAG3. METHODS Technetium-99m-MAG3 scintigraphy was performed in 20 patients who had varying degrees of renal function. Data were acquired posteriorly in supine patients at 2 sec/frame for 24 frames, 15 sec/frame for 16 frames and 30 sec/frame for 40 frames. Background correction was performed using an automated elliptical region of interest. Renal depth was estimated using improved regression equations and an empirically determined attenuation coefficient derived from phantom studies. Corrections were made for table attenuation and time discrepancies between dose injection and starting the camera. The percent injected dose in the kidney at 1-2, 1-2.5 and 2-3 min postinjection and the percent injected dose at those time periods corrected for body surface area were correlated with MAG3 clearance based on a single injection, two-compartment model. RESULTS There was high correlation between the percent injected dose in the kidney at all three time periods and the multisample clearance. Correcting for body surface areas significantly improved the correlation coefficients. Consequently, regression equations were developed to predict multisample clearance based on percent dose and body surface area. CONCLUSIONS The optimization features described in this method should improve precision when sequential studies are conducted in the same patient.