Influence of 3D mean-median filter parameters on 3D ordered subsets expectation maximization reconstructed images / 中国医学影像技术

ABSTRACT

Objective: To quantitatively analyze the influence of 3D mean-median filter parameter on ordered subsets expectation maximization (OSEM) reconstruction images through applying 3D mean-median filter to process the projection data before reconstruction and adjusting the parameter. Methods: 3D mean-median filter was applied to process 3D phantom projection data simulated by the Analytical Simulator (ASIM). OSEM algorithm of open source tomographic image reconstruction (STIR) was used to reconstruct the projection data before and after filtering. Finally, the reconstructed image was visually and quantitatively evaluated. Results: The filter parameter K was closely related to image quality. If the Kvalue was too large, the edge preservation capability of image was poor, and the image was too smooth. If the Kvalue was too small to suppress noise, the details of image were blurred. Conclusion: The noise level and edge preservation effect of the image are very sensitive to the selection of filtering parameter K. The range of filtering parameter can be selected according to the distribution of gradient histogram. The appropriate parameter can be chosen by combining with the gradient distribution ratio, so as to remove noise and retain edge characteristic.