Determining the depth distribution of planar gamma-ray sources using the shape of the Compton continuum:: Experimental test

In single photon emission computed tomography, the spatial distribution of a gamma-emitting radionuclide is often determined without taking into account the shape of the continuum of the measured γ-ray spectrum. This text considers a previously published method for using the measured Compton scattered photons to determine the depth distribution. Least-squares fitting is employed instead of kernel matrix inversion and regularisation. The method is applied to experimental measurements of a radioactive layer of 2 MBq 51Cr, using measurement times between 2 and 5 h. First, we describe the shape of the spectra, then the method used to reconstruct the depth distribution is described and tested for the almost noise free case. Furthermore, in order to assess the relation between statistical noise in the spectrum and the reliability of the reconstructed source distribution, we quantify these by defining two corresponding parameters and study their relationship. We do not consider generalisation beyond a single plane source oriented parallel to the detector face at an unknown depth, or multiples of such a source at different unknown depths.