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 Mar 4:16:1458142.
doi: 10.3389/fneur.2025.1458142. eCollection 2025.

Assessment of commercially available artificial intelligence software for differentiating hemorrhage from contrast on head CT following thrombolysis for ischemic stroke

Ayden L Olsen  1 , Daniel Thomas Ginat  2
Affiliations

Assessment of commercially available artificial intelligence software for differentiating hemorrhage from contrast on head CT following thrombolysis for ischemic stroke

Ayden L Olsen et al. Front Neurol. .

Abstract

Background: In patients who have undergone ischemic stroke therapy, retained iodine-based contrast can resemble acute intracranial hemorrhage (ICH) on standard computed tomography (CT). The purpose of this study is to determine the accuracy of commercially available artificial intelligence software for differentiating hemorrhage from contrast in such cases.

Methods: A total of 45 CT scans analyzed by Aidoc software that also included dual-energy iodine subtraction maps from dual energy CT from 23 unique patients (12 male, 11 female, age range 30-99 years, mean age 67.6 years, standard deviation 18.5 years) following recent ischemic stroke therapy were retrospectively reviewed for the presence of hemorrhage versus retained contrast material.

Results: The sensitivity and specificity of the model in detecting acute intracranial hemorrhage as opposed to contrast were 51.7 and 50.0%, respectively. The positive and negative predictive values were 65.2 and 36.4%, respectively.

Conclusion: The current Aidoc software is not optimized for differentiating between acute hemorrhage and retained contrast on CT. This justifies the development of a more robust artificial intelligence model trained to differentiate between ICH and iodine contrast based on both DECT and standard CT images.

Keywords: artificial intelligence; commercial AI; contrast CT; dual energy CT; intracranial hemorrhage.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.

Figures

Figure 1
Figure 1
Inclusion criteria for analyzed images (Created using BioRender).
Figure 2
Figure 2
ROC curve created from the Aidoc model classifications with an AUC of 0.51.
Figure 3
Figure 3
Example of a false positive reported by the Aidoc algorithm: the retained contrast in the left frontal lobe (red arrow) seen on contrast CT (left) was flagged for ICH. The iodine suppression image (right) confirms that there is no hemorrhage.
Figure 4
Figure 4
Example of a false negative reported by the Aidoc algorithm: a small area of hemorrhage along the posterior cingulate gyrus (red arrows) as seen in the contrast CT (left) is confirmed by persistence in the iodine suppression image (right).
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Yaghi S, Willey JZ, Cucchiara B, Goldstein JN, Gonzales NR, Khatri P, et al. . American Heart Association stroke council; council on cardiovascular and stroke nursing; council on clinical cardiology; and council on quality of care and outcomes research. Treatment and outcome of hemorrhagic transformation after intravenous Alteplase in acute ischemic stroke: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. (2017) 48:e343–61. doi: 10.1161/STR.0000000000000152 - DOI - PubMed
    1. Liu K, Jiang L, Ruan J, Xia W, Huang H, Niu G, et al. . The role of dual energy CT in evaluating hemorrhagic complications at different stages after thrombectomy. Front Neurol. (2020) 11:583411. doi: 10.3389/fneur.2020.583411, PMID: - DOI - PMC - PubMed
    1. Pham J, Gan C, Dabboucy J, Stella DL, Dowling R, Yan B, et al. . Occult contrast retention post-thrombectomy on 24-h follow-up dual-energy CT: associations and impact on imaging analysis. Int J Stroke. (2023) 18:1228–37. doi: 10.1177/17474930231182018, PMID: - DOI - PubMed
    1. Flohr TG, McCollough CH, Bruder H, Petersilka M, Gruber K, Süss C, et al. . First performance evaluation of a dual-source CT (DSCT) system. Eur Radiol. (2006) 16:256–68. doi: 10.1007/s00330-005-2919-2, PMID: - DOI - PubMed
    1. Ananthakrishnan L, Rajiah P, Ahn R, Rassouli N, Xi Y, Soesbe TC, et al. . Spectral detector CT-derived virtual non-contrast images: comparison of attenuation values with unenhanced CT. Abdom Radiol. (2017) 42:702–9. doi: 10.1007/s00261-016-1036-9, PMID: - DOI - PubMed

LinkOut - more resources