U-Net and YOLOv8 Artificial Intelligence Models for Automated Recognition of Internal Jugular Veins and Radial Arteries: A Foundational Study for Artificial Intelligence-guided Vascular Cannulation inPoint-of-care Ultrasound
Journal of Cardiothoracic and Vascular Anesthesia · 2026
DOI: 10.1053/j.jvca.2025.07.040 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift Journal of Cardiothoracic and Vascular Anesthesia, 2026. Verfasst von Yunxin Gu und 15 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ultrasound Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China. Electronic address: guyunxin3104@…. Hauptthemen laut MEDLINE: Artificial Intelligence, Jugular Veins, Point-of-Care Systems, Radial Artery. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 32 Literaturangaben. Gefördert von: Anhui Medical University.
Abstract (Originalsprache)
OBJECTIVES: This study compares the U-Net and You Only Look Once version 8 (YOLOv8) models for identifying internal jugular veins (IJVs) and radial arteries (RAs) in longitudinal and/or transversal ultrasound views, evaluating their vascular recognition capabilities for artificial intelligence-guided cannulation systems under point-of-care ultrasound (POCUS) visualization. DESIGN: A retrospective study. SETTING: Six teaching hospitals. PARTICIPANTS: Data from 1,122 ultrasound images (612 IJVs, 510 RAs) between January and December 2024. INTERVENTIONS: No intervention. MEASUREMENTS AND MAIN RESULTS: U-Net was employed for pixelwise segmentation with a combined dice-cross-entropy loss, while YOLOv8s-seg incorporated attention mechanisms for object detection. Model performance was evaluated using precision, recall, F1-score, mean average precision at intersection over union (IoU) threshold 0.5 (mAP@0.5), IoU, Dice coefficient, and inference time metrics. YOLOv8 demonstrated superior performance with a precision of 0.996, recall of 1.00, mAP@0.5 of 0.995, IoU of 0.739, Dice coefficient of 0.834, and inference time of 26.3 ms, outperforming U-Net, which achieved a precision of 0.988, recall of 0.998, mAP@0.5 of 0.993, IoU of 0.719, Dice coefficient of 0.816, and inference time of 38.9 ms. In the validation set, YOLOv8 exhibited higher accuracy across all categories, with values of 0.96 for V-S, 0.90 for V-L, 0.89 for A-S, and 0.86 for A-L, compared with U-Net's respective accuracies of 0.81, 0.88, 0.87, and 0.82. CONCLUSION: YOLOv8 outperformed U-Net in accuracy, precision, and speed, proving more suitable for real-time vascular localization in clinical settings. These findings provide critical algorithmic foundations for developing automated vascular puncture navigation platforms in the future.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Ultrasound Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China. Electronic address: guyunxin3104@163.com.
- Department of Ultrasound Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China. Electronic address: tang1913100032@163.com.
- Department of Ultrasound Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China.
- Department of Anesthesiology and Perioperative Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, China.
- Department of Ultrasound, Anhui Chest Hospital, Hefei, China.
- Department of Anesthesiology, the First Affiliated Hospital of Anhui University of Science & Technology, Huainan, China.
Förderung
- Anhui Medical University
Zitieren
APA
Gu, Y., Tang, G., Gao, L., Li, R., Zhang, Y., Li, R., Wu, H., Yu, L., Li, H., Han, Y., Dong, B., Wu, D., Shao, J., Chen, Y., Peng, M., & Wang, J. (2026). U-Net and YOLOv8 Artificial Intelligence Models for Automated Recognition of Internal Jugular Veins and Radial Arteries: A Foundational Study for Artificial Intelligence-guided Vascular Cannulation inPoint-of-care Ultrasound. Journal of Cardiothoracic and Vascular Anesthesia. https://doi.org/10.1053/j.jvca.2025.07.040
Vancouver
Gu Y, Tang G, Gao L, Li R, Zhang Y, Li R, et al. U-Net and YOLOv8 Artificial Intelligence Models for Automated Recognition of Internal Jugular Veins and Radial Arteries: A Foundational Study for Artificial Intelligence-guided Vascular Cannulation inPoint-of-care Ultrasound. Journal of Cardiothoracic and Vascular Anesthesia. 2026. doi:10.1053/j.jvca.2025.07.040
BibTeX
@article{Gu2026,
author = {Gu, Yunxin and Tang, Gege and Gao, Lan and Li, Rui and Zhang, Ye and Li, Rubing and Wu, Hanmei and Yu, Letao and Li, Huafeng and Han, Yang and Dong, Bingtian and Wu, Defeng and Shao, Jiaqi and Chen, Yilin and Peng, Mei and Wang, Jiajia},
title = {U-Net and YOLOv8 Artificial Intelligence Models for Automated Recognition of Internal Jugular Veins and Radial Arteries: A Foundational Study for Artificial Intelligence-guided Vascular Cannulation inPoint-of-care Ultrasound},
journal = {Journal of Cardiothoracic and Vascular Anesthesia},
year = {2026},
doi = {10.1053/j.jvca.2025.07.040}
}
EUVIMED