Purpose: A quality control (QC) system for dual-energy CT (DECT) was developed. The scope of the QC system was to monitor both the constancy of the CT images and the software used in calculating the DECT derived maps. Longitudinal analysis was based on a standard imaging protocol, a commercial multi-energy phantom, and a semi-automatic analysis tool.

Methods: The phantom consisted of an elliptical body section with round slots for interchangeable inserts. It was scanned with 90kVp/Sn150kVp, automatic tube current modulation, and 9.6 mGy CTDI. From the two conventional CT images, scanner manufacturer's software was used to provide virtual monoenergetic images at two different energies, effective atomic number (Z) maps, and iodine concentration maps. The images were analyzed using an open-source tool allowing user-selected statistics of interest. The means and standard deviations of the phantom background and the iodine, calcium, and water inserts were recorded. The QC tool is available at github.com/tomakela/dectqatool.

Results: The obtained results were generally highly consistent over time, except for the smaller diameter iodine inserts. A small change inZ was observed after a DECT software update. The developed QC tool aided the analysis robustness: the segmentations were modifiable when needed, and small rotations or air bubbles in the water insert were easily corrected.

Conclusion: The developed QC system provided easy-to-use workflow for constancy measurements. A small deviation due to change in the post-processing was detected. The proposed imaging protocol and analysis steps, and the reported measurement variations can aid in determining action levels for DECT QC.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ejmp.2024.103211DOI Listing

Publication Analysis

Top Keywords

quality control
8
semi-automatic analysis
8
imaging protocol
8
analysis
5
dual-energy computed
4
computed tomography
4
tomography quality
4
control initial
4
initial experiences
4
experiences semi-automatic
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!