Qualitative Comparison of Noncontrast Head Dual-Energy Computed Tomography Using Rapid Voltage Switching Technique and Conventional Computed Tomography.

OBJECTIVES

Dual-energy computed tomography (DECT) allows image reconstruction along a spectrum of virtual monochromatic energy levels. We sought to determine the optimal energy level(s) for viewing supratentorial brain and posterior fossa and compare those to polychromatic conventional CT (CCT). Furthermore, we compared 2 groups scanned with separate DECT imaging protocols. In addition, we quantify the radiation dose of DECT versus CCT.

METHODS

Forty-four nonacute patients underwent noncontrast DECT and recent CCT on the same scanner. Dual-energy CT images of the head were reconstructed at 5 keV intervals from 50 to 100 keV and randomized with CCT for blinded reader analysis. Evaluation of gray-white matter differentiation, posterior fossa artifact, and overall image noise was performed in consensus using a 5-point scale. In addition, the CTDIvol was compared with CCT examinations.

RESULTS

Optimal monochromatic viewing levels in evaluating gray-white matter differentiation were 50 to 55 keV and 50 to 60 keV, using regular-dose DECT (R-DECT) and low-dose DECT (L-DECT), respectively. The optimal levels for mitigating posterior fossa artifacts were 80 to 100 keV utilizing both R-DECT and L-DECT, whereas the optimal viewing levels for improved overall image noise were 60 to 65 keV and 65 to 70 keV for R-DECT and L-DECT, respectively. Readers favored both DECT techniques over CCT. The CTDIvol for DECT was 10% and 37% lower than CCT without a statistically significant reduction in image quality.

CONCLUSIONS

Optimized noncontrast DECT compared favorably to CCT, with a significant dose reduction benefit.