- Performance of Half-dose Chest Computed Tomography in Lung Malignancy Using an Iterative Reconstruction Technique
-
Hee Kang, Jung Gu Park, Se Kyoung Park, Beom Su Kim, Ki Nam Lee, Kyeung Seung Oh
-
Kosin Med J. 2017;32(1):47-57. Published online June 30, 2017
-
DOI: https://doi.org/10.7180/kmj.2017.32.1.47
-
-
Abstract
PDFPubReader ePub
-
Objectives
The purpose of this study was to evaluate the performance of half-dose chest CT using an iterative reconstruction technique in patients with lung malignancies.
Methods
The Dual-source CT scans were obtained and half-dose datasets were reconstructed with 5 different strengths in 38 adults with lung malignancies. Two radiologists graded subjective image quality; noise, contrast and sharpness at the central/peripheral lung, mediastinum and chest wall of the reconstructed half-dose images, compared with those of standard-dose images, using a three-point scale. A lesion assessment; lesion conspicuity and diagnostic confidence, was also performed. The quantitative image noises; contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) were measured and compared with those of standard-dose images.
Results
The subjective image noise in the half-dose images was less than that of the standard-dose images. The contrast in strengths 2 to 5 was superior, the sharpness of the lung parenchyma in strengths 3 to 5 was inferior, and the CNR/SNR in all strengths were higher than those of standard-dose images (P < 0.05). The improvement of subjective image noise and contrast, the decrease in sharpness, were correlated with strength level (P < 0.05). The lesion conspicuity in half-dose images of strengths 4 and 5 was decreased. The diagnostic confidence of the half-dose images of all strengths was comparable to that of the standard-dose images (P < 0.05).
Conclusions
Half-dose chest CT images using an iterative reconstruction technique show decreased image noise, increased contrast, and diagnostic confidence comparable to standard-dose images. Images reconstructed with strength 2 and 3 appear to be the optimal choice in clinical practice.
|