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
. 2023 Jul;39(7):1385-1396.
doi: 10.1007/s10554-023-02839-5. Epub 2023 Apr 7.

Coronary X-ray angiography segmentation using Artificial Intelligence: a multicentric validation study of a deep learning model

Miguel Nobre Menezes  1   2 , João Lourenço Silva  3 , Beatriz Silva  4   5 , Tiago Rodrigues  4   5 , Cláudio Guerreiro  6 , João Pedro Guedes  7 , Manuel Oliveira Santos  8   9 , Arlindo L Oliveira  3 , Fausto J Pinto  4   5
Affiliations

Coronary X-ray angiography segmentation using Artificial Intelligence: a multicentric validation study of a deep learning model

Miguel Nobre Menezes et al. Int J Cardiovasc Imaging. 2023 Jul.

Abstract

Introduction: We previously developed an artificial intelligence (AI) model for automatic coronary angiography (CAG) segmentation, using deep learning. To validate this approach, the model was applied to a new dataset and results are reported.

Methods: Retrospective selection of patients undergoing CAG and percutaneous coronary intervention or invasive physiology assessment over a one month period from four centers. A single frame was selected from images containing a lesion with a 50-99% stenosis (visual estimation). Automatic Quantitative Coronary Analysis (QCA) was performed with a validated software. Images were then segmented by the AI model. Lesion diameters, area overlap [based on true positive (TP) and true negative (TN) pixels] and a global segmentation score (GSS - 0 -100 points) - previously developed and published - were measured.

Results: 123 regions of interest from 117 images across 90 patients were included. There were no significant differences between lesion diameter, percentage diameter stenosis and distal border diameter between the original/segmented images. There was a statistically significant albeit minor difference [0,19 mm (0,09-0,28)] regarding proximal border diameter. Overlap accuracy ((TP + TN)/(TP + TN + FP + FN)), sensitivity (TP / (TP + FN)) and Dice Score (2TP / (2TP + FN + FP)) between original/segmented images was 99,9%, 95,1% and 94,8%, respectively. The GSS was 92 (87-96), similar to the previously obtained value in the training dataset.

Conclusion: the AI model was capable of accurate CAG segmentation across multiple performance metrics, when applied to a multicentric validation dataset. This paves the way for future research on its clinical uses.

Keywords: Artificial Intelligence; Coronary angiography; Coronary artery disease; Deep learning; Machine learning; Percutaneous coronary intervention..

PubMed Disclaimer

Conflict of interest statement

Not applicable.

Figures

Fig. 1
Fig. 1
Graphical Abstract: Overview of the segmentation and analysis process. Top left: Baseline CAG of a right coronary artery. Top right: AI automated segmented image. Bottom left: automatic QCA analysis image output in detail. Bottom middle: transposition of the lesion markers on the segmented image in detail. Bottom right: area overlap between the region of interest in the auto-QCA and the segmented image; white pixels are true positives; green pixels are false negatives; red pixels are false positives
Fig. 2
Fig. 2
Flowchart of patient and image selection
Fig. 3
Fig. 3
Comparative view of a right coronary artery (56% stenosis by QCA). Left-to-right: original image, auto-QCA, transposition of lines (proximal border diameter, lesion diameter and distal border diameter) to segmented image
Fig. 4
Fig. 4
Catheter segmentation assessment. Left-to-right: original image, auto-border detection by reference software, transposition of lines in proximal border to segmented image
Fig. 5
Fig. 5
Area overlap in a and left anterior descending 64% stenosis (as measured by QCA).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Wang L, Liang D, Yin X, et al. Coronary artery segmentation in angiographic videos utilizing spatial-temporal information. BMC Med Imaging 2020. 2020;201 20:1–10. doi: 10.1186/S12880-020-00509-9. - DOI - PMC - PubMed
    1. Liang D, Qiu J, Wang L, et al. Coronary angiography video segmentation method for assisting cardiovascular disease interventional treatment. BMC Med Imaging 2020. 2020;201 20:1–8. doi: 10.1186/S12880-020-00460-9. - DOI - PMC - PubMed
    1. Du T, Xie L, Zhang H, et al. Training and validation of a deep learning architecture for the automatic analysis of coronary angiography. EuroIntervention. 2021;17:32–40. doi: 10.4244/EIJ-D-20-00570. - DOI - PMC - PubMed
    1. Yang S, Kweon J, Roh J-H, et al. Deep learning segmentation of major vessels in X-ray coronary angiography. Sci Rep 2019. 2019;91 9:1–11. doi: 10.1038/s41598-019-53254-7. - DOI - PMC - PubMed
    1. Nobre Menezes M, Lourenço-Silva J, Silva B, et al. Development of deep learning segmentation models for coronary X-ray angiography: Quality assessment by a new global segmentation score and comparison with human performance. Rev Port Cardiol. 2022 doi: 10.1016/J.REPC.2022.04.001. - DOI - PubMed

LinkOut - more resources