A method was set up for single-photon emission tomographic (SPET) quantification of radioactivity concentration in small anatomical structures. The method is based on the theoretical model proposed by Kessler et al. (J. Comput Assist Tomogr 1984; 8: 514-522) describing the effects of spatial resolution (partial volume effect and spillover) on the quantification of radioactivity concentration in small spherical objects. The model was validated here in SPET, by phantom experimental measurements, in relation to object size and source/background contrast. Good agreement was found between model-predicted and SPET-measured radioactivity concentration ratios in hot spots in hot background experiments. Accuracy of the method was assessed for comparison of model-corrected and true radioactivity concentration ratios and was found to be within 8.5% over the full range of object size (9.4-36.5 mm). The good agreement found indicates that the model can be used to correct for partial volume effect and spillover in specific clinical situations, when the anatomical structure under study can be approximated by a sphere of known size (e.g. neuroreceptor and tumour studies). The method was applied to a representative SPET monoclonal antibody patient study for the quantification of radioactivity concentration in ocular melanoma.