Automated scheme of plaque classification based on segmentation in carotid ultrasound images using transformer approach
Journal of Medical Ultrasonics · 2025
DOI: 10.1007/s10396-025-01522-7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Medical Ultrasonics, 2025. Verfasst von Gakuto Hirano und 7 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Health Sciences, Fujita Health University, Toyoake, Japan. Hauptthemen laut MEDLINE: Plaque, Atherosclerotic, Carotid Arteries, Carotid Stenosis, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 38 Literaturangaben.
Abstract (Originalsprache)
PURPOSE: Carotid plaque is a major risk factor for cerebral infarction. Ultrasonography (US) is extensively used for screening carotid plaque, but US images contain more noise than those of computed tomography and magnetic resonance imaging, and the edges of the plaque regions are unclear. In addition, B-mode echogenicity evaluation, which is important for plaque risk assessment, has challenges involving the subjectivity of the evaluator. Although previous studies on carotid plaque assessment have included plaque segmentation, most studies involved manual operations. In this study, we propose an automated scheme of plaque classification based on segmentation in carotid US images using the transformer approach, to resolve the issues of previous studies and to perform plaque echogenicity classification. METHODS: The B-mode video captured in the long-axis cross-section was converted to still images, and region extraction and echogenicity classification were performed using TransUNet. The results of the TransUNet output and US images were fed into the Vision Transformer (ViT) for classification into hypoechoic or isoechoic-hyperechoic plaques. RESULTS: The Dice index, which indicates the accuracy of plaque region extraction, was 0.592. The Dice indices by echogenicity were 0.200, 0.493, and 0.542 for the hypoechoic, isoechoic, and hyperechoic regions, respectively. The balanced accuracy, which indicates the classification accuracy, was 79.6%. The correct classification rate for high-risk hypoechoic plaques was 95.2%. CONCLUSION: These results suggest that the proposed method is useful for evaluating the echogenicity classification of carotid artery plaques.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Graduate School of Health Sciences, Fujita Health University, Toyoake, Japan.
- Canon Medical Systems Corporation, Otawara, Japan.
- Faculty of Information Engineering, Meijo University, 1-501 Shiogamaguchi, Tempaku-ku, Nagoya, 468-8502, Japan. teramoto@meijo-u.ac.jp.
- Department of Radiology, Fujita Health University Hospital, Toyoake, Japan.
- Department of Clinical Laboratory, Fujita Health University Hospital, Toyoake, Japan.
- Department of Comprehensive Strokology, School of Medicine, Fujita Health University, Toyoake, Japan.
Zitieren
APA
Hirano, G., Teramoto, A., Takai, H., Sasaki, Y., Sugimoto, K., Matsumoto, S., Saito, K., & Fujita, H. (2025). Automated scheme of plaque classification based on segmentation in carotid ultrasound images using transformer approach. Journal of Medical Ultrasonics. https://doi.org/10.1007/s10396-025-01522-7
Vancouver
Hirano G, Teramoto A, Takai H, Sasaki Y, Sugimoto K, Matsumoto S, et al. Automated scheme of plaque classification based on segmentation in carotid ultrasound images using transformer approach. Journal of Medical Ultrasonics. 2025. doi:10.1007/s10396-025-01522-7
BibTeX
@article{Hirano2025,
author = {Hirano, Gakuto and Teramoto, Atsushi and Takai, Hiroji and Sasaki, Yutaka and Sugimoto, Keiko and Matsumoto, Shoji and Saito, Kuniaki and Fujita, Hiroshi},
title = {Automated scheme of plaque classification based on segmentation in carotid ultrasound images using transformer approach},
journal = {Journal of Medical Ultrasonics},
year = {2025},
doi = {10.1007/s10396-025-01522-7}
}
EUVIMED