An evaluation of four methods of 111In planar image quantification.

The accurate quantification of the in vivo distribution of 111In labeled platelets, other cells, and proteins with a scintillation camera is important in clinical and experimental medicine. Planar techniques of image quantification were therefore evaluated with the aim of improving on the accuracy, and simplifying the techniques currently in use. The attenuation of the 172- and 247-keV photons of 111In, singly and in combination, was determined for varying diameter flat sources (3.4 to 16.9 cm). The influence of region of interest (ROI) selection on the shape of the attenuation curves was also determined for five different ROI's. Defining the attenuation curves mathematically generated parameters of fit for three approaches to in vivo quantification, namely: a single exponential geometric mean approach that takes into account source size, depth-dependent, and depth-independent buildup factor approaches to account for the contribution of scatter. The accuracy of these techniques was ascertained and compared to the classical geometric mean method. This was done in a waxen phantom of a human thorax with a hollow liver and spleen. The results indicated that the depth-independent buildup factor is the best method; the error for quantification in the spleen was 0.8% +/- 2.2%. The classical geometric mean approach gave a corresponding error of 43.3% +/- 3.4%. Since the attenuation of the two energies of 111In differ, their ratio changes with depth. This phenomenon was investigated with the goal of determining whether the depth of an object can be estimated from one set of planar images. This was not successful.(ABSTRACT TRUNCATED AT 250 WORDS)