Material decomposition maps based calibration of dual energy CT scanners for proton therapy planning: a phantom study

Abstract

We introduce a new calibration method for Dual Energy CT (DECT) based on Material Decomposition (MD) maps, specifically iodine and water MD maps. The aim of this method is to provide the first DECT calibration based on MD maps. The experiments were carried out using a General Electric (GE) Revolution CT scanner with ultra-fast kV switching and used a Density Phantom by GAMMEX for calibration and evaluation.

The calibration process involves several steps. First, we tested the ability of MD values to reproduce Hounsfield Unit (HU) values of SECT acquisitions and it was found that the errors were below 1\%, validating their use for HU reproduction. Next, the different definitions of computed $Z$ values were compared and the robustness of the approach based on the materials' composition was confirmed. Finally, the calibration method was compared with a previous method by Bourque et al., providing a similar level of accuracy and superior performance in terms of precision.

Overall, this novel DECT calibration method offers improved accuracy and reliability in determining tissue-specific physical properties. The resulting maps can be valuable for proton therapy treatments, where precise dose calculations and accurate tissue differentiation are crucial for optimal treatment planning and delivery.