SU-E-J-79: Using the Distance of Closest Approach to Improve Compton Camera Image Quality

The prompt gamma radiation emitted from tissue during proton therapy offers a means of beam range verification. Compton camera (CC) imaging systems offer high efficiency and 3D imaging capability. However, Doppler broadening and the inaccuracies in the measurement of the scattering positions and energies reduce the image resolution. The purpose of our study is to determine if removing events with a distance-of-closest-approach greater than a threshold value will improve the resolution of images reconstructed using the stochastic origins ensemble (SOE) algorithm.

We first simulated a 3-stage CC detecting gammas from a 0.511 MeV point source. We then used SOE to reconstruct images from the point source and from a) all gammas, b) gammas with DCA < 3 mm, and c) gammas with DCA < 1 mm. We measured the point-spread-function for the point source for a), b), and c). Next, we simulated a 3-stage CC detecting prompt gammas emitted from tissue during proton therapy. We reconstructed the gammas using SOE and compared 2D images of all gammas, gammas with DCA < 3 mm, and gammas with DCA < 1 mm.

The FWHM of the PSF of the 0.511 MeV point source was reduced by 50% when DCA was required to be < 5 mm, and it was reduced by 65% when DCA was required to be < 3 mm. 2D images of a proton beam are of visibly higher quality as the DCA requirement is lowered.

The DCA for MC events can be used to identify the events with significant resolution loss due to the detector effects. Removing these events before running the reconstruction algorithm results in higher quality images. We discuss methods to predict the DCA based on the measured scatter data, so that a similar technique can be applied to data from real detectors. This work was supported by the National Institutes of Health through award number R21CA137362 from the National Cancer Institute.