EUVIMEDEUVIMED entdecken
Crossref

A dual-branch encoder network based on squeeze-and-excitation UNet and transformer for 3D PET-CT image tumor segmentation

Mingrui Li, Ruiming Zhu, Minghao Li, Haoran Wang, Yueyang Teng

Medical & Biological Engineering & Computing · 2026

DOI: 10.1007/s11517-025-03427-6 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical & Biological Engineering & Computing, 2026. Verfasst von Mingrui Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Medicine and Biomedical Information Engineering, Northeastern University, 110169, Shenyang, China. Hauptthemen laut MEDLINE: Imaging, Three-Dimensional, Positron Emission Tomography Computed Tomography, Neoplasms, Image Processing, Computer-Assisted. Die Arbeit nennt 47 Literaturangaben. Gefördert von: National Science Foundation of Liaoning Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recognition of tumors is very important in clinical practice and radiomics; however, the segmentation task currently still needs to be done manually by experts. With the development of deep learning, automatic segmentation of tumors is gradually becoming possible. This paper combines the molecular information from PET and the pathology information from CT for tumor segmentation. A dual-branch encoder is designed based on SE-UNet (Squeeze-and-Excitation Normalization UNet) and Transformer, 3D Convolutional Block Attention Module (CBAM) is added to skip-connection, and BCE loss is used in training for improving segmentation accuracy. The new model is named TASE-UNet. The proposed method was tested on the HECKTOR2022 dataset, which obtains the best segmentation accuracy compared with state-of-the-art methods. Specifically, we obtained results of 76.10 % and 3.27 for the two key evaluation metrics, DSC and HD95. Experiments demonstrate that the designed network is reasonable and effective. The full implementation is available at https://github.com/LiMingrui1/TASE-UNet .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Imaging, Three-DimensionalPositron Emission Tomography Computed TomographyNeoplasmsImage Processing, Computer-AssistedHumansAlgorithmsDeep Learning

Stichwörter der Autor:innen

3D medical image segmentationAttention mechanismCNNPET-CT multimodalTransformer

Autor:innen und Institutionen

  • College of Medicine and Biomedical Information Engineering, Northeastern University, 110169, Shenyang, China.
  • College of Medicine and Biomedical Information Engineering, Northeastern University, 110169, Shenyang, China. tengyy@bmie.neu.edu.cn.
  • Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, 110169, Shenyang, China. tengyy@bmie.neu.edu.cn.

Förderung

  • National Science Foundation of Liaoning Province

Zitieren

APA

Li, M., Zhu, R., Li, M., Wang, H., & Teng, Y. (2026). A dual-branch encoder network based on squeeze-and-excitation UNet and transformer for 3D PET-CT image tumor segmentation. Medical & Biological Engineering & Computing. https://doi.org/10.1007/s11517-025-03427-6

Vancouver

Li M, Zhu R, Li M, Wang H, Teng Y. A dual-branch encoder network based on squeeze-and-excitation UNet and transformer for 3D PET-CT image tumor segmentation. Medical & Biological Engineering & Computing. 2026. doi:10.1007/s11517-025-03427-6

BibTeX
@article{Li2026,
  author = {Li, Mingrui and Zhu, Ruiming and Li, Minghao and Wang, Haoran and Teng, Yueyang},
  title = {A dual-branch encoder network based on squeeze-and-excitation UNet and transformer for 3D PET-CT image tumor segmentation},
  journal = {Medical & Biological Engineering & Computing},
  year = {2026},
  doi = {10.1007/s11517-025-03427-6}
}
Kostenlos starten →