CT contrast agent evaluation model. (29th October 2018)

• Record Type: Journal Article
• Authors: Bonvento, Michael; Button, Terry; Srivastava, Pallavi
• Is Part Of: Biomedical physics & engineering express. Volume 4:Number 6(2018)
• Publication Date: 2018-10-29
• Subjects: contrast agent -- CT -- CNR -- x-ray spectra; Medical physics -- Periodicals; Biophysics -- Periodicals; Biomedical engineering -- Periodicals; Medical sciences -- Periodicals; 610.153
• DOI: 10.1088/2057-1976/aad545
• Languages: English
• ISSNs: 2057-1976

Abstract

Abstract: Contrast agents have been employed in radiography for more than 80 years and computed tomography (CT) for more than 50 years. These high atomic number agents increase photoelectric interaction thereby reducing photon intensity projected through the contrast agent containing structure to improve image contrast. A wide variety of contrast agents have been examined over the years, however, iodine and barium containing agents make up the majority of applications. Other agents may become available which may find utility due to reduced toxicity, better imaging characteristics or dose reduction. The purpose of this work is to develop a frame work to evaluate current and future CT contrast agents as they apply to CT enhancement of the head including CTA. The model used we have called the CT Contrast Agent Evaluation Model (CAEM) and allows the determination of the minimum effective concentration (MEC). The MEC is that concentration that provides a contrast-to-noise ratio (CNR) of one for a radiation dose of one centigray (cGy). The MEC can be examined as kilovoltage and filtration are varied allowing for optimized combinations. The approach taken in the model was to first compute x-ray tube spectra for the desired kilovoltage/filter combination. Next, projection data was computed through a cylindrical phantom with a central contrast containing target. Noise was applied to projection data based on photon counting. The application of filtered back projection to the noisy projection data could be used to compute a simulated image from which image CNR could be determined. This process was simplified by computing the CNR of the projection data and correcting for propagation of error using an empirically derived correction factor. The radiation exposure in air at the target can be accurately estimated from the computed x-ray spectra and used to determine dose. The computed CNR and dose were validated by direct measure using a clinical scanner.