Apr 12 2013
Image SNR with energy-selective detectors
This is the last post in my series discussing my paper, “Near optimal energy selective x-ray imaging system performance with simple detectors”. In the last post I showed plots of the signal to noise ratio (SNR) of images with different types of energy-selective detectors. In this post, I show images illustrating these differences. These images were not included in the paper but they are based on its approach. The images are calculated from a random sample of the energy spectrum at each point in a projection image. These data are then used to make images with (a) the total energy, which are comparable to the detectors now used in commercial systems, (b) the total number of photons, (c) an N2Q detector, and (d) an optimal full spectrum by weighting the spectrum data before summing, as described in Tapiovaara and Wagner (TW). I use the theory developed in my paper, to make images from A-space data using data from the N2Q detector. In order to do this, I need an estimator that achieves the Cramèr-Rao lower bound (CRLB). For this I use the A-table estimator I introduced in my paper “Estimator for photon counting energy selective x-ray imaging with multibin pulse height analysis” available for download here.
Comments Off