Investigation of the capability for material decomposition in photon counting CT images

The conventional detectors operated in the integrating mode are insensitive to the energy spectral information since the detector output is proportional to the energy fluence integrated over the whole spectrum. Recently, photon counting detectors have been considered to be an alternative device. These detectors are able to measure the x-ray photon energy deposited by each event and reduce the noise which is caused by detector leakage current using energy thresholds. Photon counting detectors with multiple energy thresholds have a significant advantage that material decomposition is possible using energy selective windows. In this study, we investigated the capability for material decomposition in images acquired from different energy window widths using the photon counting detector based on cadmium telluride (CdTe). We designed the CdTe detector and phantoms contained different materials using Geant4 Application for Tomographic Emission (GATE) simulation. To simulate the separated images of different target materials, the energy levels were chosen to be above K-edge absorption energy of target materials. For investigating the capability of material decomposition as a function of energy window width, projections were acquired from different energy windows and contrast-to-noise ratio (CNR) was calculated with constant region of interest (ROI) for target materials and PMMA in tomographic images. The CNR was increased as a function of energy window width for all target materials and decreased at above the specific energy window width. These results showed that the energy window width affects the degree of material decomposition and there are optimal window widths for target materials. In near future, we will acquire the optimal images of separated target materials using combinations of energy windows.