EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Swin Transformer and the Unet Architecture to Correct Motion Artifacts in Magnetic Resonance Image Reconstruction

Md Biddut Hossain iD, Rupali Kiran Shinde, Shariar Md Imtiaz iD, F M Fahmid Hossain, Seok-Hee Jeon, Ki-Chul Kwon iD, Nam Kim iD

International Journal of Biomedical Imaging · 02.05.2024 · 2024 · 1-12

DOI: 10.1155/2024/8972980 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Biomedical Imaging, erschienen am 02.05.2024. Verfasst von Md Biddut Hossain und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information and Communication Engineering, Chungbuk National University, Cheongju-si 28644, Chungcheongbuk-do, Republic of Korea. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Information Technology Research Center, Ministry of Science, ICT and Future Planning.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present a deep learning-based method that corrects motion artifacts and thus accelerates data acquisition and reconstruction of magnetic resonance images. The novel model, the Motion Artifact Correction by Swin Network (MACS-Net), uses a Swin transformer layer as the fundamental block and the Unet architecture as the neural network backbone. We employ a hierarchical transformer with shifted windows to extract multiscale contextual features during encoding. A new dual upsampling technique is employed to enhance the spatial resolutions of feature maps in the Swin transformer-based decoder layer. A raw magnetic resonance imaging dataset is used for network training and testing; the data contain various motion artifacts with ground truth images of the same subjects. The results were compared to six state-of-the-art MRI image motion correction methods using two types of motions. When motions were brief (within 5 s), the method reduced the average normalized root mean square error (NRMSE) from 45.25% to 17.51%, increased the mean structural similarity index measure (SSIM) from 79.43% to 91.72%, and increased the peak signal-to-noise ratio (PSNR) from 18.24 to 26.57 dB. Similarly, when motions were extended from 5 to 10 s, our approach decreased the average NRMSE from 60.30% to 21.04%, improved the mean SSIM from 33.86% to 90.33%, and increased the PSNR from 15.64 to 24.99 dB. The anatomical structures of the corrected images and the motion-free brain data were similar.

Autor:innen und Institutionen

  • Department of Information and Communication Engineering, Chungbuk National University, Cheongju-si 28644, Chungcheongbuk-do, Republic of Korea.
  • Department of Electronics Engineering, Incheon National University, 119 Academy-ro, Yeonsu-gu, Incheon 22012, Republic of Korea.

ORCID-Profile: Md Biddut Hossain, Shariar Md Imtiaz, Ki-Chul Kwon, Nam Kim

Förderung

  • Information Technology Research Center
  • Ministry of Science, ICT and Future Planning

Zitieren

APA

Hossain, M. B., Shinde, R. K., Imtiaz, S. M., Hossain, F. M. F., Jeon, S. H., Kwon, K. C., & Kim, N. (2024). Swin Transformer and the Unet Architecture to Correct Motion Artifacts in Magnetic Resonance Image Reconstruction. International Journal of Biomedical Imaging, 2024, 1-12. https://doi.org/10.1155/2024/8972980

Vancouver

Hossain MB, Shinde RK, Imtiaz SM, Hossain FMF, Jeon SH, Kwon KC, et al. Swin Transformer and the Unet Architecture to Correct Motion Artifacts in Magnetic Resonance Image Reconstruction. International Journal of Biomedical Imaging. 2024;2024:1-12. doi:10.1155/2024/8972980

BibTeX
@article{Hossain2024,
  author = {Hossain, Md Biddut and Shinde, Rupali Kiran and Imtiaz, Shariar Md and Hossain, F M Fahmid and Jeon, Seok-Hee and Kwon, Ki-Chul and Kim, Nam},
  title = {Swin Transformer and the Unet Architecture to Correct Motion Artifacts in Magnetic Resonance Image Reconstruction},
  journal = {International Journal of Biomedical Imaging},
  year = {2024},
  volume = {2024},
  pages = {1--12},
  doi = {10.1155/2024/8972980}
}
Kostenlos starten →