EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation

Rui Zhou iD, Ju Wang, Guijiang Xia, Jingyang Xing, Hongming Shen iD, Xiaoyan Shen iD

Entropy · 30.04.2024 · 26 (5) · 385

DOI: 10.3390/e26050385 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 30.04.2024. Verfasst von Rui Zhou und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Zhang Jian, Nantong University, Nantong 226019, China. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: “Six talents peaks” Project, China, Nantong Natural Science Foundation, College Students’ Innovation and Entrepreneurship Training Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In brain imaging segmentation, precise tumor delineation is crucial for diagnosis and treatment planning. Traditional approaches include convolutional neural networks (CNNs), which struggle with processing sequential data, and transformer models that face limitations in maintaining computational efficiency with large-scale data. This study introduces MambaBTS: a model that synergizes the strengths of CNNs and transformers, is inspired by the Mamba architecture, and integrates cascade residual multi-scale convolutional kernels. The model employs a mixed loss function that blends dice loss with cross-entropy to refine segmentation accuracy effectively. This novel approach reduces computational complexity, enhances the receptive field, and demonstrates superior performance for accurately segmenting brain tumors in MRI images. Experiments on the MICCAI BraTS 2019 dataset show that MambaBTS achieves dice coefficients of 0.8450 for the whole tumor (WT), 0.8606 for the tumor core (TC), and 0.7796 for the enhancing tumor (ET) and outperforms existing models in terms of accuracy, computational efficiency, and parameter efficiency. These results underscore the model’s potential to offer a balanced, efficient, and effective segmentation method, overcoming the constraints of existing models and promising significant improvements in clinical diagnostics and planning.

Schlagwörter

Stichwörter der Autor:innen

Mamba architectureMambaBTSbrain imaging segmentationcomputational complexitydice loss and cross-entropymulti-scale convolutional kernels

Autor:innen und Institutionen

  • School of Zhang Jian, Nantong University, Nantong 226019, China.
  • School of Information Science and Technology, Nantong University, Nantong 226019, China.
  • School of Microelectronics and School of Integrated Circuits, Nantong University, Nantong 226019, China.
  • Nantong Research Institute for Advanced Communication Technologies, Nantong University, Nantong 226019, China.

ORCID-Profile: Rui Zhou, Hongming Shen, Xiaoyan Shen

Förderung

  • “Six talents peaks” Project, China
  • Nantong Natural Science Foundation
  • College Students’ Innovation and Entrepreneurship Training Project

Zitieren

APA

Zhou, R., Wang, J., Xia, G., Xing, J., Shen, H., & Shen, X. (2024). Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation. Entropy, 26(5), 385. https://doi.org/10.3390/e26050385

Vancouver

Zhou R, Wang J, Xia G, Xing J, Shen H, Shen X. Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation. Entropy. 2024;26(5):385. doi:10.3390/e26050385

BibTeX
@article{Zhou2024,
  author = {Zhou, Rui and Wang, Ju and Xia, Guijiang and Xing, Jingyang and Shen, Hongming and Shen, Xiaoyan},
  title = {Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation},
  journal = {Entropy},
  year = {2024},
  volume = {26},
  number = {5},
  pages = {385},
  doi = {10.3390/e26050385}
}
Kostenlos starten →