EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

KDLM: Lightweight Brain Tumor Segmentation via Knowledge Distillation

Baotian Li, Haoyu Tang, Jun Zhu, Fang Fang Gou, Jia Wu

Advanced Intelligent Systems · 2026-05 · 8 (5)

DOI: 10.1002/aisy.202501207 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Intelligent Systems, online erschienen am 22.03.2026. Verfasst von Baotian Li und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Research Project of Guizhou Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Brain tumors pose a serious threat to human health, and accurate magnetic resonance imaging segmentation is crucial for clinical diagnosis. Existing high‐performance deep models are difficult to deploy in resource‐constrained clinical settings due to their high computational complexity. While knowledge distillation can compress models, its traditional paradigm often overlooks the heterogeneous requirements of segmentation tasks for both global context and local detail, resulting in insufficient accuracy for lightweight models. To address this issue, this study proposes a lightweight brain tumor recognition method (KDLM) based on knowledge distillation. First, we design a dual‐teacher collaboration strategy, introducing teacher models that focus on global tumor region segmentation and local detail capture, respectively, to provide complementary guidance to the student model. Second, we construct a lightweight student network (MLSNet), centered around a multiscale feature fusion module and a residual channel attention mechanism. This aims to efficiently fuse multilevel features and provide a robust network support for absorbing dual knowledge. Experiments on the MenIN and Hunan University of Medicine General Hospital glioma datasets demonstrate that the proposed method achieves Dice coefficients of 0.809 and 0.806, respectively, outperforming most existing models. This significantly reduces the number of parameters and computational cost, effectively resolving the trade‐off between accuracy and efficiency. It provides a practical solution for the implementation of AI in resource‐constrained clinical environments.

Förderung

  • Natural Science Research Project of Guizhou Province

Zitieren

APA

Li, B., Tang, H., Zhu, J., Gou, F. F., & Wu, J. (2026). KDLM: Lightweight Brain Tumor Segmentation via Knowledge Distillation. Advanced Intelligent Systems, 8(5). https://doi.org/10.1002/aisy.202501207

Vancouver

Li B, Tang H, Zhu J, Gou FF, Wu J. KDLM: Lightweight Brain Tumor Segmentation via Knowledge Distillation. Advanced Intelligent Systems. 2026;8(5). doi:10.1002/aisy.202501207

BibTeX
@article{Li2026,
  author = {Li, Baotian and Tang, Haoyu and Zhu, Jun and Gou, Fang Fang and Wu, Jia},
  title = {KDLM: Lightweight Brain Tumor Segmentation via Knowledge Distillation},
  journal = {Advanced Intelligent Systems},
  year = {2026},
  volume = {8},
  number = {5},
  doi = {10.1002/aisy.202501207}
}
Kostenlos starten →