Purpose: The effects of two dual-energy decomposition algorithms, direct decomposition and iterative decomposition, on the quality of Dual Energy Cone Beam CT (DECBCT) images are evaluated from two dimensions, accuracy of electron density (ED) and contrast-to-noise ratio (CNR), which affect dose calculation and structure identification.
Methods: By using the combination of CatPhan604 phantom and customized annulus to simulate the parts of patients of different sizes, the Edge accelerator CBCT system was used to obtain cone beam CT with two energies of 140 kVp and 100 kVp, and other parameters are the same. The material decomposition of DECBCT is carried out by using two algorithms, direct decomposition and iterative decomposition, respectively. The ED and CNR of each plug-in in the CTP682 module are calculated respectively to compare the decomposition accuracy and output image quality of the two algorithms.
Results: Based on the true value provided in the phantom manual, the ED accuracy of the two algorithms is high, and the average relative error is less than 1%, which can meet the needs of clinical dose calculation. The CNR of the iterative decomposition algorithm is significantly better than that of the direct decomposition algorithm, and the relative improvement ratio can reach up to 703.47%. Under the same scanning conditions and decomposition algorithm, large-scale phantoms will significantly reduce the ED accuracy and CNR value of DECBCT.
Conclusion: On the premise of not losing the calculation accuracy of electron density, compared with direct decomposition, iterative decomposition can reduce the image noise of the target and background areas as much as possible, thereby significantly improving the contrast, which is beneficial to provide clearer structure delineation and registration placement image.
Dual-energy Imaging, Image-guided Therapy, Cone-beam CT