An empirical material decomposition method (EMDM) for spectral CT

Material discrimination is widely used in spectral Computed Tomography (CT) with photon counting detectors (PCDs) by conducting material decomposition. Most material decomposition algorithms need to use detected spectral data. Affected by the performance of PCDs, such as charge sharing, pulse pileup, k-escape and other physical effects, it may cause serious spectra distortion that leads to errors in material decomposition and discrimination. It should be taken into account to fully take advantage of the gain from an energy-selective detector. Regarding to the complexity in modeling the physics of PCDs, we proposed a new empirical material decomposition method (EMDM) for spectral CT without the knowledge of spectral information. Using the attenuation coefficients of the calibration materials the EMDM could determine material decomposition coefficients and discriminate substances in image domain including k-edge materials by polynomial fitting under the principle of least square deviation. A series of elaborate experiments were carried out to validate the EMDM and the results show good performance in material discrimination with different concentration of different solution.