Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May 16;104(20):e42370.
doi: 10.1097/MD.0000000000042370.

Reduction of artifacts from contrast media in spectral-detector CT by combined use of virtual monoenergetic images and orthopedic metal-artifact-reduction algorithm

Affiliations

Reduction of artifacts from contrast media in spectral-detector CT by combined use of virtual monoenergetic images and orthopedic metal-artifact-reduction algorithm

Nannan Pan et al. Medicine (Baltimore). .

Abstract

This study assesses the artifacts reduction caused by contrast media (CM) in the subclavian and axillary veins in dual-layer spectral-detector CT using virtual monoenergetic images (VMI) and orthopedic metal-artifact-reduction (O-MAR) algorithm. A total of 61 nonconsecutive patients who underwent enhanced chest CT examinations were enrolled in the present study. Conventional images (CI), VMI, CI with O-MAR algorithm (CI + O-MAR), and VMI with O-MAR algorithm (VMI + O-MAR) were reconstructed using the same arterial CT dataset. The VMI and VMI + O-MAR images were reconstructed at 130 keV. Subjective image quality assessment was performed in terms of degree of artifacts and overall image quality using the Likert-scale. The differences in image noise, artifact index (AI) and CT number difference were compared among the 4 reconstructed images. Soft tissue adjacent to CM exhibited fewer artifacts and better image quality in VMI + O-MAR images than in VMI, CI + O-MAR images, and CI (P < .001). In addition, image noise and AI in VMI + O-MAR images were lower than those in VMI and CI + O-MAR images (P < .001). CT number difference was lower in VMI and VMI + O-MAR images than in CI + O-MAR images (P < .01). The CI had the highest values in image noise, AI and CT number difference (P < .001). The combination of O-MAR and 130 keV VMI showed a significant reduction of artifacts from CM than each technique alone and provided better image quality and diagnostic confidence. The combined application of O-MAR and 130 keV VMI can be a better alternative to reduce artifacts from CM in spectral-detector CT.

Keywords: artifacts; contrast media; orthopedic metal-artifact-reduction algorithm; spectral-detector CT; virtual monoenergetic images.

PubMed Disclaimer

Conflict of interest statement

The authors have no funding and conflicts of interest to disclose.

Figures

Figure 1.
Figure 1.
Location of ROIs for the objective image analysis. ROIs were drawn in pectoralis major muscle and pectoralis minor muscle close to the CM and most affected by artifacts (red ROIs). ROI was drawn in pectoralis major muscle on the contralateral side which it was not affected by artifacts (blue ROI). Furthermore, ROI was drawn in the artifact-free subcutaneous fat layer (green ROI). ROIs (40–60 mm²) were placed on CI and copied to VMI, CI + O-MAR and VMI + O-MAR images to guarantee the same size, shape and slice. CI = conventional images, CM = contrast media, O-MAR = orthopedic metal-artifact-reduction, ROIs = regions of interest, VMI = virtual monoenergetic images.
Figure 2.
Figure 2.
Bar charts showing the image noise (A), AI (B), and CT number difference (C) in the 4 reconstructed images (CI, VMI, CI + O-MAR, and VMI + O-MAR). For the image noise, all pairs showed significant differences (P < .001). There were significant differences in AI among the 4 reconstructed images (P < .001), except between VMI and CI + O-MAR image (P < .446). There were significant differences in CT number difference (P ≤ .006), except between VMI and VMI + O-MAR images (P = .069). AI = artifact index, CI = conventional images, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.
Figure 3.
Figure 3.
Subjective image quality scores of the 4 reconstructed images (CI, VMI, CI + O-MAR, and VMI + O-MAR). VMI + O-MAR showed the lowest degree of artifacts (A) and the best overall image quality (B), and had more images that were graded 1 and 2. The scores of VMI, CI + O-MAR, and VMI + O-MAR were significantly lower than CI. CI = conventional images, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.
Figure 4.
Figure 4.
A 74-year-old man with lung cancer. Enhanced chest CT images show an irregular mass in the right lung near the mediastinum. On CI (A), it is hard to evaluate the intralesional density and the relationship with mediastinum due to the artifacts from CM in the superior vena cava (arrows). The VMI (B) and CI + O-MAR (C) reduce some artifacts and improve the image quality. The VMI + O-MAR image (D) shows a mass with homogenous density that has not invaded the mediastinum. Subjective assessments from both readers generated the best scores for the VMI + O-MAR image. CI = conventional images, CM = contrast media, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.
Figure 5.
Figure 5.
An 83-year-old woman with a soft tissue mass in the right upper lobe. Enhanced chest CT images show a soft tissue mass with irregular margin and homogeneous density (circles). On CI (A), it is difficult to assess the mass because of the severe artifacts due to CM in the subclavian vein. VMI (B) reduces image artifacts to some extent, but some remaining artifacts are still visible. The CI + O-MAR image (C) reduces the artifacts, but some new streak artifacts are observed. However, the VMI + O-MAR image (D) reduces most of the artifacts and enables the evaluation of the shape, margin, and density of the mass. CI = conventional images, CM = contrast media, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.
Figure 6.
Figure 6.
A 65-year-old woman with lung cancer. Enhanced chest CT images show pleural effusion and pleural thickening (circles). Artifacts caused by CM in the superior vena cava are the most severe on CI (A). VMI (B), and CI + O-MAR image (C) reduced some artifacts. VMI + O-MAR image (D) obtains the best subjective assessment scores from both reviewers. CI = conventional images, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.
Figure 7.
Figure 7.
A 54-year-old man with an axillary lymph node. On CI (A), the axillary lymph node (long arrow) and adjacent soft tissue (short arrow) are difficult to evaluate because of the artifacts caused by CM in the subclavian vein. The VMI (B) and CI + O-MAR image (C) reduce some artifacts, but photo starvation artifacts and new streak artifacts are observed on VMI and CI + O-MAR image, respectively. The artifacts around the soft tissue are further reduced and the lymph node is clearly depicted in the VMI + O-MAR image (D). CI = conventional images, CM = contrast media, O-MAR = orthopedic metal-artifact-reduction, VMI = virtual monoenergetic images.

Similar articles

References

    1. Fleischmann D. Use of high-concentration contrast media in multiple-detector-row CT: principles and rationale. Eur Radiol. 2003;13(Suppl 5):M14–20. - PubMed
    1. Kul M, Kuru Öz D, Gürsoy Coruh A, et al. . Biphasic split-bolus injection protocol for routine contrast-enhanced chest CT: comparison with conventional early-phase single bolus technique. Br J Radiol. 2022;95:20210775. - PMC - PubMed
    1. Mitsumori LM, Wang E, May JM, et al. . Triphasic contrast bolus for whole-chest ECG-gated 64-MDCT of patients with nonspecific chest pain: evaluation of arterial enhancement and streak artifact. AJR Am J Roentgenol. 2010;194:W263–271. - PubMed
    1. McEvoy SH, Halpenny DF, Viteri-Jusué A, et al. . Investigation of patterns of nodal metastases in BRAF mutant lung cancer. Lung Cancer. 2017;108:62–5. - PMC - PubMed
    1. Shimizu Y, Koike T, Hasebe T, et al. . Surgical treatment outcomes of patients with non-small cell lung cancer and lymph node metastases. Cancers. 2023;15:3098. - PMC - PubMed