Multi-Materials Decomposition using clinical Dual-energy CT

Dual energy computed tomography(DECT) can give the variation of the attenuation coefficient with energy. Furthermore, the material information of the scanned object can be deduced. In medical application of DECT, the scanned body can be seen as a combination of several materials, such as bone, soft tissue, fat and blood. A two-steps multi-materials decomposition method was proposed to attain the composition information. Firstly, a projection domain decomposition is performed to estimate the attenuation coefficient function of the scanned object. Then the estimated attenuation coefficient function is decomposed with the attenuation coefficient functions of basis materials. There is a problem that in the first step, the spectrum of the x-ray source or additional calibration data is necessary but they may not be available in practice. We proposed an image domain multi-materials decomposition method. The basic framework is the same with the two-steps multi-materials decomposition method. The difference occurs in the first step that a direct reconstruction to dual energy datasets not the projection domain decomposition is conducted. Accordingly, the attenuation coefficients of basis materials for decomposition in the second step should be carefully selected or constructed because the direct reconstructed value is the effective attenuation coefficient. The attenuation coefficients of basis materials are chosen from a density-based clustering to the dual energy reconstruction data. We did some experiments based on clinical DECT data. Without knowing exact spectrum, the material information acquired by the proposed method is relatively accurate.