EUVIMEDEUVIMED entdecken
Crossref

MultiTrans: Multi-branch transformer network for medical image segmentation

Yanhua Zhang, Gabriella Balestra, Ke Zhang, Jingyu Wang, Samanta Rosati, Valentina Giannini

Computer Methods and Programs in Biomedicine · 2024

DOI: 10.1016/j.cmpb.2024.108280 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2024. Verfasst von Yanhua Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electronics and Telecommunications, Politecnico di Torino, Corso Duca degli Abruzzi 24, Turin, 10129, Italy; School of Astronautics, No…. Hauptthemen laut MEDLINE: Neural Networks, Computer, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 68 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVE: Transformer, which is notable for its ability of global context modeling, has been used to remedy the shortcomings of Convolutional neural networks (CNN) and break its dominance in medical image segmentation. However, the self-attention module is both memory and computational inefficient, so many methods have to build their Transformer branch upon largely downsampled feature maps or adopt the tokenized image patches to fit their model into accessible GPUs. This patch-wise operation restricts the network in extracting pixel-level intrinsic structural or dependencies inside each patch, hurting the performance of pixel-level classification tasks. METHODS: To tackle these issues, we propose a memory- and computation-efficient self-attention module to enable reasoning on relatively high-resolution features, promoting the efficiency of learning global information while effective grasping fine spatial details. Furthermore, we design a novel Multi-Branch Transformer (MultiTrans) architecture to provide hierarchical features for handling objects with variable shapes and sizes in medical images. By building four parallel Transformer branches on different levels of CNN, our hybrid network aggregates both multi-scale global contexts and multi-scale local features. RESULTS: MultiTrans achieves the highest segmentation accuracy on three medical image datasets with different modalities: Synapse, ACDC and M&Ms. Compared to the Standard Self-Attention (SSA), the proposed Efficient Self-Attention (ESA) can largely reduce the training memory and computational complexity while even slightly improve the accuracy. Specifically, the training memory cost, FLOPs and Params of our ESA are 18.77%, 20.68% and 74.07% of the SSA. CONCLUSIONS: Experiments on three medical image datasets demonstrate the generality and robustness of the designed network. The ablation study shows the efficiency and effectiveness of our proposed ESA. Code is available at: https://github.com/Yanhua-Zhang/MultiTrans-extension.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerImage Processing, Computer-AssistedHumansAlgorithmsDiagnostic ImagingDatabases, Factual

Stichwörter der Autor:innen

Abdominal multi-organ segmentationCardiac segmentationDeep learningEfficient self-attentionMedical image segmentationMulti-branch transformers

Autor:innen und Institutionen

  • Department of Electronics and Telecommunications, Politecnico di Torino, Corso Duca degli Abruzzi 24, Turin, 10129, Italy; School of Astronautics, Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 71007…
  • Department of Electronics and Telecommunications, Politecnico di Torino, Corso Duca degli Abruzzi 24, Turin, 10129, Italy. Electronic address: gabriella.balestra@polito.it.
  • School of Astronautics, Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 710072, China. Electronic address: zhangke@nwpu.edu.cn.
  • School of Astronautics, Northwestern Polytechnical University, 127 West Youyi Road, Xi'an, 710072, China. Electronic address: jywang@nwpu.edu.cn.
  • Department of Electronics and Telecommunications, Politecnico di Torino, Corso Duca degli Abruzzi 24, Turin, 10129, Italy. Electronic address: samanta.rosati@polito.it.
  • Department of Surgical Sciences, University of Turin, Turin, 10124, Italy; Radiology Unit, Candiolo Cancer Institute, FPO-IRCCS, Candiolo, 10060, Italy. Electronic address: valentina.giannini@unito.it.

Zitieren

APA

Zhang, Y., Balestra, G., Zhang, K., Wang, J., Rosati, S., & Giannini, V. (2024). MultiTrans: Multi-branch transformer network for medical image segmentation. Computer Methods and Programs in Biomedicine. https://doi.org/10.1016/j.cmpb.2024.108280

Vancouver

Zhang Y, Balestra G, Zhang K, Wang J, Rosati S, Giannini V. MultiTrans: Multi-branch transformer network for medical image segmentation. Computer Methods and Programs in Biomedicine. 2024. doi:10.1016/j.cmpb.2024.108280

BibTeX
@article{Zhang2024,
  author = {Zhang, Yanhua and Balestra, Gabriella and Zhang, Ke and Wang, Jingyu and Rosati, Samanta and Giannini, Valentina},
  title = {MultiTrans: Multi-branch transformer network for medical image segmentation},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2024},
  doi = {10.1016/j.cmpb.2024.108280}
}
Kostenlos starten →