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 1;15(5):4515-4526.
doi: 10.21037/qims-24-1949. Epub 2025 Apr 16.

Deep learning-based key point detection algorithm assisting vessel centerline extraction

Xiqian Zhang #  1   2 , Wanqing Sun #  3 , Hui Zhang #  4 , Long Yang  1 , Xiong Yang  3 , Yufei Mao  3 , Chengcheng Zhu  5 , Zhang Shi  6 , Jia Gu  7 , Juan Pan  8 , Guanxun Cheng  4 , Xin Liu  1   2   9   10 , Fei Feng  4 , Na Zhang  1   2   9   10
Affiliations

Deep learning-based key point detection algorithm assisting vessel centerline extraction

Xiqian Zhang et al. Quant Imaging Med Surg. .

Abstract

Background: Vessel centerline extraction assists in the quantitative analysis of plaque. Current algorithms generate significant errors for tortuous vessels, leading to inaccurate centerline extraction. This study proposed a key point detection algorithm to assist in vessel centerline extraction for the further quantitative analysis of plaque.

Methods: A total of 539 patients with cerebrovascular disease from multiple centers were enrolled in this retrospective study. All the patients underwent 3.0-T magnetic resonance imaging (MRI) scans. Based on the experimental experience of radiologists and clinical requirements, 32 key points were chosen, including the carotid siphon, tiny vessels, and vessel bifurcations. Accurate point detection can improve the accuracy of centerline detection. The evaluation indices included the number of undetected points (undetected_num), the number of erroneously detected points (errodetected_num), and the accuracy of each point (pointacc). The average centerline distance (ACD) was used to evaluate the improvement in centerline extraction.

Results: The average accuracy of the algorithm in detecting of the 32 points was 88.99%, and the algorithm had an accuracy exceeding 90% for 18 of these points. The accuracy of the algorithm at the sharp bend of the carotid siphon section reached 97%. The accuracy of the algorithm in detecting the points in the internal carotid artery and middle cerebral artery was 95.4%. Using the key point detection algorithm, the ACD for the right carotid artery was reduced to 0.484±0.321 mm but was 0.529±0.334 mm without the key point detection algorithm. The time required to detect the 32 key points was reduced from 319.843±6.434 to 2.046±0.315 seconds when the algorithm was used.

Conclusions: The proposed algorithm was able to automatically and accurately detect the 32 key points, especially those in the internal carotid artery and middle cerebral artery, improving vessel centerline extraction accuracy, and thus assisting in plaque assessment.

Keywords: Key point detection; magnetic resonance vessel wall imaging (MR-VWI); vessel centerline.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-1949/coif). W.S., X.Y., and Y.M. report that they are employed by Shanghai United Imaging Healthcare Co., Ltd. The other authors have no conflicts of interest to declare.

Figures

Figure 1
Figure 1
Key points and sampling range. (A) The 32 points selected for the experiment. (B) Sampling range. BA, basilar artery; LACA, left anterior cerebral artery; LCCA, left common carotid artery; LECA, left external carotid artery; LICA, left internal carotid artery; LMCA, left middle cerebral artery; LPCA, left posterior cerebral artery; LVA, left vertebral artery; RACA, right anterior cerebral artery; RCCA, right common carotid artery; RECA, right external carotid artery; RICA, right internal carotid artery; RMCA, right middle cerebral artery; RPCA, right posterior cerebral artery; RVA, right vertebral artery.
Figure 2
Figure 2
Flowchart of the experiment.
Figure 3
Figure 3
Diagram of the V-Net network structure.
Figure 4
Figure 4
The MSE and MSE2 loss. (A) The lack of convergence of the loss function during the training epochs with MSE. (B) The convergence behavior of the loss function during the training epochs with MSE2. The green line represents the validation loss, the orange line represents the training loss, and the red line represents the fitting curve of the training loss. MSE, mean squared error.
Figure 5
Figure 5
Results of centerline extraction with and without key point detection. (A) Vessel extraction with and without key point detection. The yellow arrows indicate the anterior curvature of the carotid siphon; the red marks indicate the detected vessel. (B) Vessel centerline extraction at the carotid siphon with and without ley point detection. The orange arrows indicate the carotid siphon. (C) The ACD with and without key point detection. ACD, average centerline distance; LICA, left internal carotid artery; RICA, right internal carotid artery.
Figure 6
Figure 6
Comparison of the time required for the manual and automatic key point localization.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Feigin VL, Brainin M, Norrving B, Martins S, Sacco RL, Hacke W, Fisher M, Pandian J, Lindsay P. World Stroke Organization (WSO): Global Stroke Fact Sheet 2022. Int J Stroke 2022;17:18-29. 10.1177/17474930211065917 - DOI - PubMed
    1. Baradaran H, Gupta A. Extracranial Vascular Disease: Carotid Stenosis and Plaque Imaging. Neuroimaging Clin N Am 2021;31:157-66. 10.1016/j.nic.2021.02.002 - DOI - PMC - PubMed
    1. Ambrose JA, Srikanth S. Vulnerable plaques and patients: improving prediction of future coronary events. Am J Med 2010;123:10-6. 10.1016/j.amjmed.2009.07.019 - DOI - PubMed
    1. Saba L, Saam T, Jäger HR, Yuan C, Hatsukami TS, Saloner D, Wasserman BA, Bonati LH, Wintermark M. Imaging biomarkers of vulnerable carotid plaques for stroke risk prediction and their potential clinical implications. Lancet Neurol 2019;18:559-72. 10.1016/S1474-4422(19)30035-3 - DOI - PubMed
    1. Saba L, Brinjikji W, Spence JD, Wintermark M, Castillo M, de Borst GJ, et al. Roadmap Consensus on Carotid Artery Plaque Imaging and Impact on Therapy Strategies and Guidelines: An International, Multispecialty, Expert Review and Position Statement. AJNR Am J Neuroradiol 2021;42:1566-75. 10.3174/ajnr.A7223 - DOI - PMC - PubMed

LinkOut - more resources