Daniele Marin MD, Presenter: Nothing to Disclose

Achille Mileto MD, Abstract Co-Author: Nothing to Disclose

Juan Carlos Ramirez Giraldo PhD, Abstract Co-Author: Employee, Siemens AG

To investigate the impact of a second-generation virtual monochromatic algorithm on the conspicuity of hypervascular liver tumors and image noise, using dual-source dual-energy MDCT.

A custom anthropomorphic liver phantom simulating different levels of enhancement of hypervascular lesions in three adult body sizes was imaged with a second-generation dual-source MDCT using both dual-energy (100/Sn140 kVp) and single-energy acquisitions, at various energy levels (80,100,120, 140 kVp). For each phantom size, the radiation output was kept constant for all scans. Virtual monochromatic images from the dual-energy dataset were reconstructed at energy levels ranging from 40 to 140 keV, using both first-generation (Syngo DE Monoenergetic) and second-generation (Syngo DE Monoenergetic PLUS) virtual monochromatic algorithms. Noise and tumor-to-liver contrast-to-noise ratio (CNR) were calculated and compared among different reconstructed datasets, for all phantom body sizes.

On single-energy imaging, the minimum noise level was observed at 120 kVp for the small and at 140 kVp for the medium and large phantom sizes; 80 kVp yielded the highest tumor-to-liver CNR for all phantom sizes. For the first-generation virtual monochromatic algorithm, noise was lowest at 70 keV in the small and medium phantom sizes, and 80 keV in the large phantom size; an energy level of 60 keV yielded the highest tumor-to-liver CNR for all phantom sizes. For the second-generation virtual monochromatic algorithm, noise was lowest at 80 keV for the small and the medium phantom sizes, and 90 keV for the large phantom size; an energy level of 40 keV yielded the highest tumor-to-liver CNR for all phantom sizes. In the large phantom, second-generation virtual monochromatic images at an optimal energy yielded significantly higher tumor-to-liver CNR, compared to either single-energy or first-generation virtual monochromatic images (P<0.01).

Second-generation virtual monochromatic algorithm may improves the conspicuity of hypervascular liver tumors compared to single-energy and first-generation virtual monochromatic images, in larger body sizes.

Second-generation optimal energy virtual monochromatic images may substantially improve the conspicuity of hypervascular liver tumors in larger patients.

Marin, D, Mileto, A, Ramirez Giraldo, J, Impact of a Second-Generation Virtual Monochromatic Algorithm on the Conspicuity of Hypervascular Liver Tumors Using Dual-Source Dual-Energy MDCT.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14015381.html