Window selection for dual photopeak window scatter correction in Tc-99m imaging

The width and placement of the windows for the dual photopeak window (DPW) scatter subtraction method for Tc-99m imaging is investigated in order to obtain a method that is stable on a multihead detector system for single photon emission computed tomography (SPECT) and is capable of providing a good scatter estimate for extended objects. For various window pairs, stability and noise were examined with experiments using a SPECT system, while Monte Carlo simulations were used to predict the accuracy of scatter estimates for a variety of objects and to guide the development of regression relations for various window pairs. The DPW method that resulted from this study was implemented with a symmetric 20% photopeak window composed of a 15% asymmetric photopeak window and a 5% lower window abutted at 7 keV below the peak. A power function regression was used to relate the scatter-to-total ratio to the lower window-to-total ratio at each pixel, from which an estimated scatter image was calculated. DPW demonstrated good stability, achieved by abutting the two windows away from the peak. Performance was assessed and compared with Compton window subtraction (CWS). For simulated extended objects, DPW generally produced a less biased scatter estimate than the commonly used CWS method with k = 0.5. In acquisitions of a clinical SPECT phantom, contrast recovery was comparable for both DPW and CWS; however, DPW showed greater visual contrast in clinical SPECT bone studies