U-Net and Its Variants Based Automatic Tracking of Radial Artery in Ultrasonic Short-Axis Views: A Pilot Study

Overview

Journal Diagnostics (Basel)

Specialty Radiology

Date 2024 Nov 9

PMID 39518327

Authors

Yuan Tian

Ruiyang Gao

Xinran Shi

Jiaxin Lang

Yang Xue

Chunrong Wang

Yuelun Zhang

Le Shen

Chunhua Yu

Zhuhuang Zhou

Affiliations

Soon will be listed here.

Abstract

Radial artery tracking (RAT) in the short-axis view is a pivotal step for ultrasound-guided radial artery catheterization (RAC), which is widely employed in various clinical settings. To eliminate disparities and lay the foundations for automated procedures, a pilot study was conducted to explore the feasibility of U-Net and its variants in automatic RAT. Approved by the institutional ethics committee, patients as potential RAC candidates were enrolled, and the radial arteries were continuously scanned by B-mode ultrasonography. All acquired videos were processed into standardized images, and randomly divided into training, validation, and test sets in an 8:1:1 ratio. Deep learning models, including U-Net and its variants, such as Attention U-Net, UNet++, Res-UNet, TransUNet, and UNeXt, were utilized for automatic RAT. The performance of the deep learning architectures was assessed using loss functions, dice similarity coefficient (DSC), and Jaccard similarity coefficient (JSC). Performance differences were analyzed using the Kruskal-Wallis test. The independent datasets comprised 7233 images extracted from 178 videos of 135 patients (53.3% women; mean age: 41.6 years). Consistent convergence of loss functions between the training and validation sets was achieved for all models except Attention U-Net. Res-UNet emerged as the optimal architecture in terms of DSC and JSC (93.14% and 87.93%), indicating a significant improvement compared to U-Net (91.79% vs. 86.19%, < 0.05) and Attention U-Net (91.20% vs. 85.02%, < 0.05). This pilot study validates the feasibility of U-Net and its variants in automatic RAT, highlighting the predominant performance of Res-UNet among the evaluated architectures.

References

Cao J, Chen J, Gu Y, Liu J . MFA-UNet: a vessel segmentation method based on multi-scale feature fusion and attention module. Front Neurosci. 2023; 17:1249331. PMC: 10702608. DOI: 10.3389/fnins.2023.1249331. View

Li J, Liu K, Hu Y, Zhang H, Heidari A, Chen H . Eres-UNet++: Liver CT image segmentation based on high-efficiency channel attention and Res-UNet+. Comput Biol Med. 2023; 158:106501. DOI: 10.1016/j.compbiomed.2022.106501. View

Zeng C, Zhao G, Deng J, Li D . Oblique versus longitudinal axis/in-plane approaches for ultrasound-guided radial arterial cannulation: A randomised controlled trial. Eur J Anaesthesiol. 2020; 37(7):618-621. DOI: 10.1097/EJA.0000000000001186. View

Wang Z, Guo H, Shi S, Xu Y, Ye M, Bai L . Long-axis in-plane combined with short-axis out-of-plane technique in ultrasound-guided arterial catheterization in infants: A randomized controlled trial. J Clin Anesth. 2023; 85:111038. DOI: 10.1016/j.jclinane.2022.111038. View

Zhou Z, Siddiquee M, Tajbakhsh N, Liang J . UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Deep Learn Med Image Anal Multimodal Learn Clin Decis Support (2018). 2020; 11045:3-11. PMC: 7329239. DOI: 10.1007/978-3-030-00889-5_1. View

Cao X, Chen H, Li Y, Peng Y, Wang S, Cheng L . Uncertainty Aware Temporal-Ensembling Model for Semi-Supervised ABUS Mass Segmentation. IEEE Trans Med Imaging. 2020; 40(1):431-443. DOI: 10.1109/TMI.2020.3029161. View

Imbriaco G, Monesi A, Giugni A, Cilloni N . Radial artery cannulation in intensive care unit patients: Does distance from wrist joint increase catheter durability and functionality?. J Vasc Access. 2020; 22(4):561-567. DOI: 10.1177/1129729820953020. View

Ding W, Wang J, Zhou W, Zhou S, Chang C, Shi J . Joint Localization and Classification of Breast Cancer in B-Mode Ultrasound Imaging via Collaborative Learning With Elastography. IEEE J Biomed Health Inform. 2022; 26(9):4474-4485. DOI: 10.1109/JBHI.2022.3186933. View

Imbriaco G, Monesi A, Spencer T . Preventing radial arterial catheter failure in critical care - Factoring updated clinical strategies and techniques. Anaesth Crit Care Pain Med. 2022; 41(4):101096. DOI: 10.1016/j.accpm.2022.101096. View

10.

Franco-Sadud R, Schnobrich D, Mathews B, Candotti C, Abdel-Ghani S, Perez M . Recommendations on the Use of Ultrasound Guidance for Central and Peripheral Vascular Access in Adults: A Position Statement of the Society of Hospital Medicine. J Hosp Med. 2019; 14(9):E1-E22. PMC: 10193861. DOI: 10.12788/jhm.3287. View

11.

Zolfaghari H, Andiapen M, Baumbach A, Mathur A, Kerswell R . Wall shear stress and pressure patterns in aortic stenosis patients with and without aortic dilation captured by high-performance image-based computational fluid dynamics. PLoS Comput Biol. 2023; 19(10):e1011479. PMC: 10635572. DOI: 10.1371/journal.pcbi.1011479. View

12.

Vila M, Remeseiro B, Grau M, Elosua R, Betriu A, Fernandez-Giraldez E . Semantic segmentation with DenseNets for carotid artery ultrasound plaque segmentation and CIMT estimation. Artif Intell Med. 2020; 103:101784. DOI: 10.1016/j.artmed.2019.101784. View

13.

Schmidt G, Blaivas M, Conrad S, Corradi F, Koenig S, Lamperti M . Ultrasound-guided vascular access in critical illness. Intensive Care Med. 2019; 45(4):434-446. DOI: 10.1007/s00134-019-05564-7. View

14.

Azzopardi C, Camilleri K, Hicks Y . Bimodal Automated Carotid Ultrasound Segmentation Using Geometrically Constrained Deep Neural Networks. IEEE J Biomed Health Inform. 2020; 24(4):1004-1015. DOI: 10.1109/JBHI.2020.2965088. View

15.

Saruwatari M, Nguyen T, Fooladi Talari H, Matisoff A, Sharma K, Donoho K . Assessing the Effect of Augmented Reality on Procedural Outcomes During Ultrasound-Guided Vascular Access. Ultrasound Med Biol. 2023; 49(11):2346-2353. PMC: 10658651. DOI: 10.1016/j.ultrasmedbio.2023.07.011. View

16.

Pacha H, Alahdab F, Al-Khadra Y, Idris A, Rabbat F, Darmoch F . Ultrasound-guided versus palpation-guided radial artery catheterization in adult population: A systematic review and meta-analysis of randomized controlled trials. Am Heart J. 2018; 204:1-8. DOI: 10.1016/j.ahj.2018.06.007. View

17.

Ashkani Chenarlogh V, Shabanzadeh A, Oghli M, Sirjani N, Farzin Moghadam S, Akhavan A . Clinical target segmentation using a novel deep neural network: double attention Res-U-Net. Sci Rep. 2022; 12(1):6717. PMC: 9038725. DOI: 10.1038/s41598-022-10429-z. View

18.

Szarski M, Chauhan S . Improved real-time segmentation of Intravascular Ultrasound images using coordinate-aware fully convolutional networks. Comput Med Imaging Graph. 2021; 91:101955. DOI: 10.1016/j.compmedimag.2021.101955. View

19.

Wang J, Chen B, Gao F . Exploring hemodynamic mechanisms and re-intervention strategies for partial false lumen thrombosis in Stanford type B aortic dissection after thoracic aortic endovascular repair. Int J Cardiol. 2024; 417:132494. DOI: 10.1016/j.ijcard.2024.132494. View

20.

Bhattacharjee S, Maitra S, Baidya D . Comparison between ultrasound guided technique and digital palpation technique for radial artery cannulation in adult patients: An updated meta-analysis of randomized controlled trials. J Clin Anesth. 2018; 47:54-59. DOI: 10.1016/j.jclinane.2018.03.019. View