Optimizing dual-energy CT technique for iodine-based contrast-to-noise ratio

TL;DR

This paper presents a systematic method for optimizing dual-energy CT parameters to maximize iodine contrast-to-noise ratio by analyzing noise propagation and deriving optimal settings for tube voltage and virtual monochromatic energy.

Contribution

The study introduces an analytic framework for selecting optimal DECT scan parameters to enhance image quality, including a novel approach to determine the best virtual monochromatic energy and tube settings.

Findings

Optimal VMI energy for maximal CNR is the least noisy synthesized monochromatic image.

Tube potential pair of 60/120 kV yields maximal CNR at 55 KeV VMI.

Using more general weights in image combination does not improve image quality.

Abstract

Purpose: This study proposes a systematic method for determining the optimal x-ray tube settings/energy windows and fluence for minimal noise and maximum CNR in material density images obtained from DECT scans by fixing the subject size and the total radiation dose. Methods: The noise propagation in the process of sinogram and image reconstruction from DECT measurements is analyzed. Analytic estimates for the sinogram and monochromatic image pixel variances and the CNR as functions of tube potentials, fluence, and virtual monochromatic image (VMI) energy are derived, and then used in a phantom experiment as an objective function for optimizing the tube settings to minimize the image noise and maximize the CNR. Results: A non-trivial example that shows the existence of singular solutions to the inversion of sinograms-to-DECT measurements map was presented. Additionally, the optimal VMI energy for maximal CNR was determined. The optimal energy VMI was found to be the least noisy monochromatic image synthesized from the iodine and water density images, and it was shown that using more general weights in combining the two images linearly does not improve image quality. When the x-ray beam filter material was fixed at 2mm of Aluminum and the photon fluence for low and high kV scans were considered equal, the tube potential pair of 60/120 kV led to the maximal CNR in the VMI formed at energy 55 KeV. Conclusions: Optimizing DECT scan parameters to maximize the CNR can be done in a systematic way. Also choosing the parameters that maximize the Jacobian determinant over the sinogram domain would lead to more stable reconstructions due to the reduced amplification of the measurement noise. Since the values of the Jacobian determinant depend strongly on the imaging task, careful consideration of all of the relevant factors is needed when implementing the proposed framework.