Author: Izabella L. Barreto, Aroon Pressram 👨🔬
Affiliation: University of Florida College of Medicine, University of Florida 🌍
Purpose: Our clinical dual energy CT (DECT) head protocol reconstructs iodine-suppressed virtual non-contrast (VNC) images using the vendor-recommended slope for material discrimination. However, this protocol led to inadequate suppression of high iodine concentrations in phantom images, raising concerns about clinical accuracy. This study aims to adapt the slope to achieve accurate iodine suppression and ensure more reliable DECT imaging.
Methods: A multi-energy phantom was scanned on a sequential DECT scanner using a head protocol with 135/80 kVp and 100/570 mA. Basis material virtual monoenergetic images (VMIs) were reconstructed at 52 keV and 66 keV. For each iodine concentration (0.2 mg/ml – 20 mg/ml), CT numbers were measured with ROIs and plotted for low (52 keV) and high (66 keV) energies to determine a measured slope. Using the NIST X-COM database, CT numbers were calculated for all iodine concentrations, which were plotted for low and high energies to calculate a theoretical slope. VNC images were reconstructed using the vendor-recommended slope, the measured slope, and the theoretical slope. Iodine suppression in VNC images was compared by measuring CT number differences between iodine concentrations and the phantom’s solid water background.
Results: The measured slope (0.59) matched the theoretical slope (0.59), but differed from the vendor-recommended slope (0.55). For higher iodine concentrations (10,15, 20 mg/ml), VNC images reconstructed with the vendor slope showed unsuppressed iodine, incorrectly appearing as blood, with higher CT number differences from background (69.2, 111.1, 136.1 HU, respectively) compared to the measured and theoretical slopes (4.7, 14.7, 6.5 HU, respectively). The measured and theoretical slopes demonstrated superior iodine suppression, minimizing residual iodine signal in VNC images.
Conclusion: Adapting the material discrimination slope to measured values led to improved suppression of high concentrations of iodine in VNC images. Institutions should independently assess material mapping parameters to ensure reliable DECT imaging.