Characterization of Noise and Noise Correlation in Photon-Counting CT for Multimaterial Decomposition-Based Spectral Imaging

Photon-counting CT offers the capability of spectral imaging to meet the challenges imposed by advanced clinical application. It has been well understood that noise and its correlation in material-specific (basis) images play a vital role in determining the performance of spectral imaging based on two-material decomposition. Recently, to further understand the basics of spectral imaging, the property of noise and its correlation in the material-specific images obtained via multimaterial decomposition (m-MD) has been investigated for spectral imaging in photon-counting CT, in which the equations that govern the behavior of noise and noise correlation in the material-specific (basis) images in m-MD are derived through statistical analysis. Taking ideal and realistic detector spectral response into consideration, the derived equations have been evaluated and verified via simulation studies that run over the scenarios of three- and four-material decomposition (3-MD and 4-MD, that is, m-MD). The obtained results are believed to be informative to the community of researchers who are interested in spectral imaging in photon-counting CT, including scientists in academia, engineers in industry, and clinicians in clinics. Moreover, the characterization of noise and its correlation of m-MD based spectral imaging is not only for photon-counting CT but also extendable to energy-integration CT and other X-ray-related imaging modalities. Hence, aimed at helping understand the fundamentals that are essential to the implementation and optimization of multimaterial decomposition-based spectral imaging in photon-counting CT, energy-integration CT, and other X-ray imaging modalities, the results of our recent investigation are briefly presented in this chapter.