Evaluation of the quantitative capability of a home-made cone-beam micro computed tomography system.

Quantitative computed tomography provides the most accurate form of bone mineral density measurement. It is very useful in several applications when used with micro computed tomography (microCT). The objective of this study is to evaluate the quantitative capability of iterative and analytic reconstruction with and without energy-based beam hardening calibration using a home-made microCT. Due to the fact that the source of X-rays in the microCT is poly-energetic and the linear attenuation coefficient varies with the energy of the X-ray photons, a specific correction is presented in this study that resolves the poly-energetic effect. Then, a 3D distribution of the linear attenuation coefficient is reconstructed from the ordered subsets using the maximum likelihood and T-FDK algorithms. These algorithms were both developed for cone-beam microCT. The images reconstructed by the two algorithms with/without correction are presented. A region of interest analysis is used to evaluate the results from two algorithms and their advantages and disadvantages are discussed. The quantitative capability is better when the image is reconstructed using the iterative method along with a beam hardening correction.