Quantification of technetium-99m lung radioactivity from planar images

Six methods of quantifying technetium-99m lung uptake from planar gamma camera images were evaluated. Camera sensitivities, the broad beam attenuation coefficient and build-up factors were derived from suitable phantom measurements. The accuracy of the methods was evaluated by quantifying the lung uptake of 99mTc macroaggregated albumin (MAA) in ten patients and assuming complete trapping by the lung of the known activity of injected MAA particles. Three methods based on published techniques which related the count rate from an image of a lung to that from a lung phantom were the least accurate, producing lung activities which were typically about 70% of the injected activity. Of the other three techniques, the depth-dependent build-up factor method was slightly more accurate than the geometric mean and the depth-independent build-up factor methods, producing average values (+/- SD) of lung activity which were 100% +/- 3%, 106% +/- 3% and 101% +/- 5%, respectively, of the injected activity. To measure lung uptake, all of these latter three methods required an attenuation correction with a flood source transmission scan, and therefore their accuracy was affected by the variation in the activity distribution and attenuation across an image of the thorax.