A novel Klein-Nishina based scatter correction method for SPECT and planar imaging

An algorithm correcting for the fraction of scattered events in SPECT and planar images has been developed. The algorithm utilises a pixel-based multi-channel analyser for data acquisition and works locally, pixel-by-pixel or on clusters of pixels. The differential Klein-Nishina cross-section, modified to fit the energy resolution and the sensitivity of a specific gamma camera was first determined. Furthermore, in a selected energy window, covering part of the Compton distribution, the modified Klein-Nishina distribution was scaled to the same count level as the experimental spectra. The scaled Klein-Nishina distribution was subsequently used for estimating the amount of scattered photons in the upper half of the photo-peak. To ensure a stable peak position a pixel peak-alignment routine was used. After subtraction of the estimated scatter distribution the number of unscattered photons are given. Assuming the photo-peak to be symmetrical in absence of scatter, the unscattered photon distribution in the upper half of the photo-peak may be mirrored (folded) into the lower part of the photo-peak, thereby estimating the "clean" photo-peak in that part of the window.