EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A dual-branch and dual attention transformer and CNN hybrid network for ultrasound image segmentation

Chong Zhang, Lingtong Wang, Guohui Wei, Zhiyong Kong, Min Qiu

Frontiers in Physiology · 27.09.2024 · 15

DOI: 10.3389/fphys.2024.1432987 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Physiology, online erschienen am 27.09.2024. Verfasst von Chong Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Intelligence and Information Engineering, Shandong University of Traditional Chinese Medicine, Jinan, China. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Ultrasound imaging has become a crucial tool in medical diagnostics, offering real-time visualization of internal organs and tissues. However, challenges such as low contrast, high noise levels, and variability in image quality hinder accurate interpretation. To enhance the diagnostic accuracy and support treatment decisions, precise segmentation of organs and lesions in ultrasound image is essential. Recently, several deep learning methods, including convolutional neural networks (CNNs) and Transformers, have reached significant milestones in medical image segmentation. Nonetheless, there remains a pressing need for methods capable of seamlessly integrating global context with local fine-grained information, particularly in addressing the unique challenges posed by ultrasound images.

Methods

In this paper, to address these issues, we propose DDTransUNet, a hybrid network combining Transformer and CNN, with a dual-branch encoder and dual attention mechanism for ultrasound image segmentation. DDTransUNet adopts a Swin Transformer branch and a CNN branch to extract global context and local fine-grained information. The dual attention comprising Global Spatial Attention (GSA) and Global Channel Attention (GCA) modules to capture long-range visual dependencies. A novel Cross Attention Fusion (CAF) module effectively fuses feature maps from both branches using cross-attention.

Results

Experiments on three ultrasound image datasets demonstrate that DDTransUNet outperforms previous methods. In the TN3K dataset, DDTransUNet achieves IoU, Dice, HD95 and ACC metrics of 73.82%, 82.31%, 16.98 mm, and 96.94%, respectively. In the BUS-BRA dataset, DDTransUNet achieves 80.75%, 88.23%, 8.12 mm, and 98.00%. In the CAMUS dataset, DDTransUNet achieves 82.51%, 90.33%, 2.82 mm, and 96.87%.

Discussion

These results indicate that our method can provide valuable diagnostic assistance to clinical practitioners.

Schlagwörter

Stichwörter der Autor:innen

attention mechanismconvolutional neural networkdeep learningtransformerultrasound image segmentation

Autor:innen und Institutionen

  • School of Intelligence and Information Engineering, Shandong University of Traditional Chinese Medicine, Jinan, China.
  • Department of Ultrasound Medicine, Affiliated Hospital of Shandong University of Traditional Chinese Medicine, Jinan, China.
  • Thyroid Surgery, Jining Medical University, Jining, China.

Zitieren

APA

Zhang, C., Wang, L., Wei, G., Kong, Z., & Qiu, M. (2024). A dual-branch and dual attention transformer and CNN hybrid network for ultrasound image segmentation. Frontiers in Physiology, 15. https://doi.org/10.3389/fphys.2024.1432987

Vancouver

Zhang C, Wang L, Wei G, Kong Z, Qiu M. A dual-branch and dual attention transformer and CNN hybrid network for ultrasound image segmentation. Frontiers in Physiology. 2024;15. doi:10.3389/fphys.2024.1432987

BibTeX
@article{Zhang2024,
  author = {Zhang, Chong and Wang, Lingtong and Wei, Guohui and Kong, Zhiyong and Qiu, Min},
  title = {A dual-branch and dual attention transformer and CNN hybrid network for ultrasound image segmentation},
  journal = {Frontiers in Physiology},
  year = {2024},
  volume = {15},
  doi = {10.3389/fphys.2024.1432987}
}
Kostenlos starten →