In this paper, we demonstrate how to form images of D2-dopamine receptor concentration in the brain with positron emission tomography (PET). The signal-to-noise ratio (SNR) of the reconstructed dynamic PET verileri is generally so low that region of interest (ROI) analysis is required to have accurate quantification. ROI analysis produces a single value of receptor concentration for each region in the brain. Elimination of intermediate reconstruction steps and direct reconstruction of parameters from sinograms allowed us to form dense images of parameters of interest . In this paper, we applied the direct parametric reconstruction algorithm to form images of D2-dopamine receptor concentration within monkey brain using 18F-fallypride radiotracer. Our initial investigations show that our D2 receptor images are consistent with the results of the classical ROI analysis.