Research on Low-Dose CT Image Denoising Methods Based on Self-Supervised Learning
Frontiers in Computing and Intelligent Systems · 30.04.2026 · 16 (2) · 30-36
DOI: 10.54097/1924aq22 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Computing and Intelligent Systems, online erschienen am 30.04.2026. Verfasst von An Ge. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Low-dose CT images are affected by noise and artifacts, compromising diagnostic reliability. While deep learning-based denoising methods have shown promise, most rely on paired normal-dose CT images, which are difficult to obtain clinically. To address this, this study proposes a self-supervised denoising method using pseudo-labels generated by guided image filtering, along with an Attention-Gate-enhanced REsidual Denoising network (AG-REDCNN). Experiments demonstrate that the method effectively reduces noise and preserves structural details without requiring normal-dose CT images during training, outperforming several existing approaches.
Zitieren
APA
Ge, A. (2026). Research on Low-Dose CT Image Denoising Methods Based on Self-Supervised Learning. Frontiers in Computing and Intelligent Systems, 16(2), 30-36. https://doi.org/10.54097/1924aq22
Vancouver
Ge A. Research on Low-Dose CT Image Denoising Methods Based on Self-Supervised Learning. Frontiers in Computing and Intelligent Systems. 2026;16(2):30-36. doi:10.54097/1924aq22
BibTeX
@article{Ge2026,
author = {Ge, An},
title = {Research on Low-Dose CT Image Denoising Methods Based on Self-Supervised Learning},
journal = {Frontiers in Computing and Intelligent Systems},
year = {2026},
volume = {16},
number = {2},
pages = {30--36},
doi = {10.54097/1924aq22}
}
EUVIMED