K-edge digital subtraction imaging with dichromatic x-ray sources: SNR and dose studies.

The aim of the present work is to analytically evaluate the signal to noise ratio (SNR) and the delivered dose in K-edge digital subtraction imaging (KES) using two types of x-ray sources: a monochromatic x-ray source (available at synchrotron radiation facilities and considered as gold standard) and a quasi-monochromatic compact source. The energy separation DeltaE between the two monochromatic beams is 1 keV and 4 keV for the two sources, respectively. The evaluation has been performed for both radiography and computed tomography. Different geometries have been studied to mimic clinical situations. In mammography, a pathology perfused by a contrast agent has been modelled; in angiography, a vessel superimposed to a ventricle or a stand-alone artery stenosis has been studied. The SNR and the skin dose have been calculated as a function of the detail diameter, the contrast agent (iodine and gadolinium), and its concentration in the tissues. Results show that for DeltaE = 4 keV a slightly higher delivered dose is required to obtain the same SNR with respect to DeltaE < 1 keV. A similar study has been performed for KES-CT. Computer simulations of CT images performed with Snark software are shown to validate the analytical calculations.