Room: Exhibit Hall | Forum 9
Purpose: To assess Hounsfield unit (HU) variability due to beam hardening in a dual energy CT (DECT) system. Algorithms used for reconstructing iodine maps and virtual non-contrast (VNC) image series do not use the same beam hardening correction as routine studies, leading to potential streaking and cupping beam hardening artifacts. We aim to measure differences in HUs in regions where cupping and streaking artifacts present.
Methods: A tissue equivalent multi-energy CT (MECT) head phantom was scanned using a DECT system. Calcium inserts (50, 100, 300 mg/mL) were inserted in the MECT phantom to mimic calcifications and simulate beam hardening conditions. Three regions of interest (ROI) at equal distance from the center of the phantom were placed where streaking artifacts presented, and in regions without streaking artifacts. HU values were measured in ROIs at the periphery and center of the phantom to quantify the cupping artifact. The reported measurement error accounted for photon statistics/electronic noise, scanner reproducibility, and positioning reproducibility.
Results: The solid water periphery was mis-characterized as containing iodine due to reconstruction artifact: the iodine map enhanced the periphery and the VNC suppressed the periphery. A cupping artifact resulting in a 26% change in HU value between the phantom center and periphery was seen in the iodine map, and an inverse cupping artifact showing a 21% HU change was seen in the VNC image. In comparison to regions without streaking artifacts, regions with streaking artifacts demonstrated differences of up to -7 % HU and -13% HU for the iodine map and VNC, respectively.
Conclusion: HU variability due to beam hardening may cause material differentiation errors in the iodine maps and VNCs. We recommend for vendors to implement beam-hardening correction into DECT reconstruction algorithms.
Funding Support, Disclosures, and Conflict of Interest: Pending funding from CANON USA: Title of of the award submitted: Clinical and Physics Evaluation of Dual Energy CT No conflicts of interest to report