EfficientUNetViT: Efficient Breast Tumor Segmentation Utilizing UNet Architecture and Pretrained Vision Transformer
Bioengineering · 21.09.2024 · 11 (9) · 945
DOI: 10.3390/bioengineering11090945 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioengineering, online erschienen am 21.09.2024. Verfasst von Shokofeh Anari und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Accounting, Economic and Financial Sciences, Islamic Azad University, South Tehran Branch, Tehran 1584743311, Iran. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study introduces a sophisticated neural network structure for segmenting breast tumors. It achieves this by combining a pretrained Vision Transformer (ViT) model with a UNet framework. The UNet architecture, commonly employed for biomedical image segmentation, is further enhanced with depthwise separable convolutional blocks to decrease computational complexity and parameter count, resulting in better efficiency and less overfitting. The ViT, renowned for its robust feature extraction capabilities utilizing self-attention processes, efficiently captures the overall context within images, surpassing the performance of conventional convolutional networks. By using a pretrained ViT as the encoder in our UNet model, we take advantage of its extensive feature representations acquired from extensive datasets, resulting in a major enhancement in the model’s ability to generalize and train efficiently. The suggested model has exceptional performance in segmenting breast cancers from medical images, highlighting the advantages of integrating transformer-based encoders with efficient UNet topologies. This hybrid methodology emphasizes the capabilities of transformers in the field of medical image processing and establishes a new standard for accuracy and efficiency in activities related to tumor segmentation.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Accounting, Economic and Financial Sciences, Islamic Azad University, South Tehran Branch, Tehran 1584743311, Iran.
- Poli.TIC-CTI-Renato Archer, Campinas 13069-901, Brazil.
- School of Computing, Faculty of Engineering and Computing, Dublin City University, D09 V209 Dublin, Ireland.
- School of Electrical and Computer Engineering, State University of Campinas, Campinas 13083-852, Brazil.
- ADAPT Research Centre, School of Computer Science, University of Galway, H91 TK33 Galway, Ireland.
ORCID-Profile: Shokofeh Anari, Gabriel Gomes de Oliveira, Ramin Ranjbarzadeh, Angela Maria Alves, Gabriel Caumo Vaz, Malika Bendechache
Zitieren
APA
Anari, S., de Oliveira, G. G., Ranjbarzadeh, R., Alves, A. M., Vaz, G. C., & Bendechache, M. (2024). EfficientUNetViT: Efficient Breast Tumor Segmentation Utilizing UNet Architecture and Pretrained Vision Transformer. Bioengineering, 11(9), 945. https://doi.org/10.3390/bioengineering11090945
Vancouver
Anari S, de Oliveira GG, Ranjbarzadeh R, Alves AM, Vaz GC, Bendechache M. EfficientUNetViT: Efficient Breast Tumor Segmentation Utilizing UNet Architecture and Pretrained Vision Transformer. Bioengineering. 2024;11(9):945. doi:10.3390/bioengineering11090945
BibTeX
@article{Anari2024,
author = {Anari, Shokofeh and de Oliveira, Gabriel Gomes and Ranjbarzadeh, Ramin and Alves, Angela Maria and Vaz, Gabriel Caumo and Bendechache, Malika},
title = {EfficientUNetViT: Efficient Breast Tumor Segmentation Utilizing UNet Architecture and Pretrained Vision Transformer},
journal = {Bioengineering},
year = {2024},
volume = {11},
number = {9},
pages = {945},
doi = {10.3390/bioengineering11090945}
}
EUVIMED