EUVIMEDEUVIMED entdecken
Crossref

Efficient vision mamba for MRI super‐resolution via hybrid selective scanning

Mojtaba Safari, Shansong Wang, Vanessa L Wildman, Mingzhe Hu, Zach Eidex, Chih-Wei Chang, Erik H Middlebrooks, Richard L J Qiu, Pretesh Patel, Ashesh B Jani, Hui Mao, Zhen Tian und 1 weitere

Medical Physics · 2026-05 · 53 (5)

DOI: 10.1002/mp.70508 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Medical Physics, online erschienen am 27.05.2026. Verfasst von Mojtaba Safari und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Image Processing, Computer-Assisted, Deep Learning, Signal-To-Noise Ratio. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: NIH HHS, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

High‐resolution MRI is essential for accurate diagnosis and treatment planning, but its clinical acquisition is often constrained by long scanning times, which increase patient discomfort and reduce scanner throughput. While super‐resolution (SR) techniques offer a post‐acquisition solution to enhance resolution, existing deep learning approaches face trade‐offs between reconstruction fidelity and computational efficiency, limiting their clinical applicability.

Purpose

This study aims to develop an efficient and accurate deep learning framework for MRI SR that preserves fine anatomical detail while maintaining low computational overhead, enabling practical integration into clinical workflows. Materials and Methods We propose a novel SR framework based on multi‐head selective state‐space models (MHSSM) integrated with a lightweight channel multilayer perceptron (MLP). The model employs 2D patch extraction with hybrid scanning strategies (vertical, horizontal, and diagonal) to capture long‐range dependencies while mitigating pixel forgetting. Each MambaFormer block combines MHSSM, depthwise convolutions, and gated channel mixing to balance local and global feature representation. The framework was trained and evaluated on two distinct datasets: 7T brain T1 MP2RAGE maps (142 subjects) and 1.5T prostate T2w MRI (334 subjects). Performance was compared against multiple baselines including Bicubic interpolation, GAN‐based (CycleGAN, Pix2pix, SPSR), transformer‐based (SwinIR), Mamba‐based (MambaIR), and diffusion‐based (I 2 SB, Res‐SRDiff) methods.

Results

The proposed model demonstrated superior performance across all evaluation metrics while maintaining exceptional computational efficiency. On the 7T brain dataset, our method achieved the highest structural similarity (SSIM: ) and peak signal‐to‐noise ratio (PSNR: dB), along with the best perceptual quality scores (LPIPS: ; GMSD: ). These results represented statistically significant improvements over all baselines (), including a 2.1% SSIM gain over SPSR and a 2.4% PSNR improvement over Res‐SRDiff. For the prostate dataset, the model similarly outperformed competing approaches, achieving SSIM of , PSNR of dB, LPIPS of , and GMSD of . Notably, our framework accomplished these results with only 0.9 million parameters and 57 GFLOPs, representing reductions of 99.8% in parameters and 97.5% in computational operations compared to Res‐SRDiff, while also substantially outperforming SwinIR and MambaIR in both accuracy and efficiency metrics.

Conclusion

The proposed framework provides a computationally efficient yet accurate solution for MRI SR, delivering well‐defined anatomical details and improved perceptual fidelity across anatomically distinct datasets. By significantly reducing computational demands while maintaining state‐of‐the‐art performance, the model offers strong potential for feasibility toward clinical translation and scalable integration into future imaging workflows.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Magnetic Resonance ImagingImage Processing, Computer-AssistedDeep LearningSignal-To-Noise RatioMultilayer PerceptronsHumansProstate

Stichwörter der Autor:innen

MRISSMdeep learningstate‐space modelsuper‐resolutionultra‐high field MRI

Autor:innen und Institutionen

  • Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.
  • Department of Radiation and Cellular Oncology, University of Chicago, Chicago, Illinois, USA.
  • Department of Radiology, Mayo Clinic, Jacksonville, Florida, USA.
  • Department of Radiology and Imaging Science and Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.

Förderung

  • NIH HHS
  • National Institutes of Health

Zitieren

APA

Safari, M., Wang, S., Wildman, V. L., Hu, M., Eidex, Z., Chang, C. W., Middlebrooks, E. H., Qiu, R. L. J., Patel, P., Jani, A. B., Mao, H., Tian, Z., & Yang, X. (2026). Efficient vision mamba for MRI super‐resolution via hybrid selective scanning. Medical Physics, 53(5). https://doi.org/10.1002/mp.70508

Vancouver

Safari M, Wang S, Wildman VL, Hu M, Eidex Z, Chang CW, et al. Efficient vision mamba for MRI super‐resolution via hybrid selective scanning. Medical Physics. 2026;53(5). doi:10.1002/mp.70508

BibTeX
@article{Safari2026,
  author = {Safari, Mojtaba and Wang, Shansong and Wildman, Vanessa L and Hu, Mingzhe and Eidex, Zach and Chang, Chih-Wei and Middlebrooks, Erik H and Qiu, Richard L J and Patel, Pretesh and Jani, Ashesh B and Mao, Hui and Tian, Zhen and Yang, Xiaofeng},
  title = {Efficient vision mamba for MRI super‐resolution via hybrid selective scanning},
  journal = {Medical Physics},
  year = {2026},
  volume = {53},
  number = {5},
  doi = {10.1002/mp.70508}
}
Kostenlos starten →