To evaluate the ability of retrospectively generated virtual monoenergetic images (VMIs) from the detector-based spectral computed tomography (SDCT) to augment pulmonary artery enhancement in CT and if iodine map can predict the optimal monoenergetic level.
The study included 79 patients with contrast-enhanced chest CT scans on an SDCT scanner. Conventional 120-kVp images and VMI from 40 to 80 keV were generated. Attenuation, noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured at 7 different locations in the pulmonary arterial system. The iodine concentration (in milligrams per milliliter) was calculated using the iodine-density images. The overall image quality was subjectively graded on a 5-point scale, with 1 being the worst and 5 the best. Fifty-four patients with suboptimal pulmonary enhancement (<200 Hounsfield units [HU]) were then identified. From the VMIs, an ideal set was chosen that maintained mean vascular attenuation greater than 200 HU while maintaining at least diagnostically acceptable quality (ie, IQ score ≥3). At this ideal energy level, quantitative and qualitative parameters were compared with the standard 120-kVp polyenergetic study. Average iodine concentrations were correlated with the optimal keV levels used for salvaging suboptimal studies.
The mean attenuation of all the measured pulmonary arterial regions in the suboptimal cases was 136.1 ± 18.1 HU in conventional 120-kVp images. Attenuations of the VMIs at 40, 50, and 60 keV were significantly higher than conventional images measuring 357.5 ± 19.5, 243.6 ± 16.7, and 176.6 ± 15.0 HU, respectively (P < 0.001). Similar results were seen with SNR and CNR. In total, 50 studies can be salvaged, with 50 keV being the optimal energy for 21, 60 keV optimal for 17, and 40 keV optimal for 12 studies. At the optimal energy level, there were improvements of attenuation, SNR, and CNR by 71%, 63%, and 137% compared with conventional images. There was a positive correlation between iodine value and optimal reconstruction energy with a linear equation y = 5.9539x + 27.434 and R2 = 0.8093.
Suboptimal enhanced pulmonary arterial CT studies can be salvaged using low-energy VMI generated from the SDCT scanner. There were significant improvements of attenuation, SNR, and CNR at the optimal monoenergetic level.
From the *Department of Radiology, University Hospitals Cleveland Medical Center; and
†Department of Radiology, Cleveland Clinic Medical Foundation, Cleveland, OH; and
‡Department of Radiology, Texas Children's Hospital, Houston, TX;
§Philips Healthcare, Cleveland, OH; and
∥Cardiothoracic Imaging, Department of Radiology, UT Southwestern Medical Center, Dallas, TX.
Received for publication November 6, 2017; accepted February 15, 2018.
Correspondence to: Prabhakar Rajiah, MBBS, MD, FRCR, Department of Radiology, UT Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390 (e-mail: Prabhakar.Rajiah@utsouthwestern.edu).
The study was funded by institutional research grant from Philips Healthcare.
A.D. is an employee of Philips Healthcare. P.R. has received in the past (>2 years back) honoraria from Philips Healthcare.
The other authors declare no conflict of interest.