EUVIMEDEUVIMED entdecken
Open Access

ZS-KAN: Zero-Shot Image Denoising With Lightweight Kolmogorov--Arnold Networks

Jianxu Wang, Ge Wang

IEEE Transactions on Biomedical Engineering · 2026-07 · 73 (7) · 2602-2611

DOI: 10.1109/tbme.2025.3640551 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Biomedical Engineering, 2026-07. Verfasst von Jianxu Wang und 1 weiteren Autor:innen. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Neural Networks, Computer. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NIBIB NIH HHS, NHLBI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

OBJECTIVE: Current learning-based image denoising methods have achieved impressive performance. However, their reliance on deep neural architectures and large datasets limits their applicability in data-limited or edge-computing scenarios. Therefore, we aim to develop a lightweight, effective, and efficient neural network for self-supervised image denoising. METHODS: Motivated by the functional approximation capability of Kolmogorov-Arnold networks (KANs), here we present ZS-KAN, a zero-shot denoising approach built upon a hybrid architecture that combines the computational efficiency of convolutional neural networks with the representational flexibility of KANs. RESULTS: Experiments on synthetic and real-world noisy data show that ZS-KAN achieves performance comparable to or even surpassing state-of-the-art zero-shot denoising methods while using only 1% -25% of their parameter counts, significantly reducing model complexity. Visual comparisons indicate that ZS-KAN typically preserves more fine details than competing methods. CONCLUSION: This study demonstrates a strong potential of KANs as a powerful component for zero-shot image denoising. SIGNIFICANCE: The combination of lightweight architecture, self-learning, and computational efficiency leads to the merits of ZS-KAN for real-world deployment, particularly in medical imaging workflows.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Processing, Computer-AssistedNeural Networks, ComputerAlgorithmsConvolutional Neural NetworksHumansSignal-To-Noise Ratio

Förderung

  • NCI NIH HHS
  • NIBIB NIH HHS
  • NHLBI NIH HHS
  • National Institutes of Health

Zitieren

APA

Wang, J., & Wang, G. (2026). ZS-KAN: Zero-Shot Image Denoising With Lightweight Kolmogorov--Arnold Networks. IEEE Transactions on Biomedical Engineering, 73(7), 2602-2611. https://doi.org/10.1109/tbme.2025.3640551

Vancouver

Wang J, Wang G. ZS-KAN: Zero-Shot Image Denoising With Lightweight Kolmogorov--Arnold Networks. IEEE Transactions on Biomedical Engineering. 2026;73(7):2602-2611. doi:10.1109/tbme.2025.3640551

BibTeX
@article{Wang2026,
  author = {Wang, Jianxu and Wang, Ge},
  title = {ZS-KAN: Zero-Shot Image Denoising With Lightweight Kolmogorov--Arnold Networks},
  journal = {IEEE Transactions on Biomedical Engineering},
  year = {2026},
  volume = {73},
  number = {7},
  pages = {2602--2611},
  doi = {10.1109/tbme.2025.3640551}
}
Kostenlos starten →