Multiscale frequency attention transformer for resolution enhancement in magnetic particle imaging.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος:	Multiscale frequency attention transformer for resolution enhancement in magnetic particle imaging.
Συγγραφείς:	Jiang W; School of Control Science and Engineering, Shandong University, Jinan, Shandong, China., Xu J; School of Control Science and Engineering, Shandong University, Jinan, Shandong, China., Yang X; Weifang Xinli Superconducting Magnet Technology Co., Ltd, Weifang, China., He N; School of Life Science and Technology, Xidian University, Xi'an, Shaanxi, China., Wildgruber M; Klinikum der Universität München, Munich, Germany., Zhong J; School of Instrumentation and Opto-Electronic Engineering, and Ministry of Education Key Laboratory of Precision Opto-Mechatronics Technology, Beihang University, Beijing, China., Zhao J; Children's Hospital Affiliated to Shandong University, Jinan, Shandong, China., Ma X; School of Control Science and Engineering, Shandong University, Jinan, Shandong, China.
Πηγή:	Medical physics [Med Phys] 2026 Aug; Vol. 53 (8), pp. e70565.
Τύπος έκδοσης:	Journal Article
Γλώσσα:	English
Στοιχεία περιοδικού:	Publisher: John Wiley and Sons, Inc Country of Publication: United States NLM ID: 0425746 Publication Model: Print Cited Medium: Internet ISSN: 2473-4209 (Electronic) Linking ISSN: 00942405 NLM ISO Abbreviation: Med Phys Subsets: MEDLINE
Imprint Name(s):	Publication: 2017- : Hoboken, NJ : John Wiley and Sons, Inc. Original Publication: Lancaster, Pa., Published for the American Assn. of Physicists in Medicine by the American Institute of Physics.
Ιατρικοί όροι (MeSH):	Magnetic Particle Imaging/methods , Image Processing, Computer-Assisted/methods , Deep Learning*, Signal-To-Noise Ratio
Περίληψη:	Background: Magnetic particle imaging (MPI) is an emerging functional imaging modality that enables high-resolution (HR) visualization of superparamagnetic iron oxide nanoparticles (SPIONs). It offers significant advantages, including high penetration capability, absence of ionizing radiation, high contrast, and exceptional temporal resolution and sensitivity, making it highly promising for a broad spectrum of biomedical and clinical applications. Purpose: Although MPI holds great promise, fundamental spatial resolution is inherently constrained by the gradient field strength and the saturation magnetization physics of the SPIONs. To address the image blurring characterized by the system's point spread function (PSF) and the dependency on high-gradient fields for HR imaging, this study aims to develop a robust deep learning framework to reduce reliance on high gradient hardware. We aim to achieve high-fidelity resolution enhancement computationally, recovering fine structural details from low-resolution inputs, thereby bypassing the need for costly hardware upgrades. Methods: We propose a Multiscale Frequency Attention Transformer (MFAT) model, which integrates a frequency-domain discriminative feedforward network (DFFN) with a multiscale attention block (MAB) to form an end-to-end framework for recovering HR MPI images from low-resolution reconstructions. The MFAT architecture combines multiscale feature aggregation with a novel frequency attention mechanism, allowing it to capture both global context and fine local details across multiple scales while enhancing fidelity and sharpness in the frequency domain. Results: Extensive evaluations using both simulated and real-world MPI data demonstrate that the proposed MFAT model outperforms existing state-of-the-art deep learning methods, achieving an improvement of 7.51% in peak signal-to-noise ratio (PSNR) in the MNIST dataset, corresponding to a gain of 2.48 dB over the second strongest baseline. Furthermore, in terms of spatial resolution, MFAT significantly reduced the full width at half maximum (FWHM) by 17.86% on the complex FIVES vascular structures compared to the second-performing baseline, verifying its superior capability in recovering high-frequency details. Conclusions: Thus, MFAT provides an effective solution for enhancing spatial resolution in MPI image reconstruction and significantly improves image quality. This advancement holds strong potential to facilitate the translation of MPI technology into more refined clinical applications. (© 2026 American Association of Physicists in Medicine.)
References:	Curr Opin Chem Biol. 2018 Aug;45:131-138. (PMID: 29754007) Sci Rep. 2017 Jul 31;7(1):6872. (PMID: 28761103) Adv Mater. 2024 Apr;36(17):e2306450. (PMID: 37812831) IEEE Trans Med Imaging. 2020 Jun;39(6):2133-2139. (PMID: 31940524) Phys Med Biol. 2023 Feb 10;68(4):. (PMID: 36689774) AJNR Am J Neuroradiol. 2019 Feb;40(2):206-212. (PMID: 30655254) Med Phys. 2025 Aug;52(8):e18036. (PMID: 40781733) Med Image Anal. 2024 Oct;97:103280. (PMID: 39096845) Phys Med Biol. 2020 Sep 24;65(19):195004. (PMID: 32764190) Nature. 2005 Jun 30;435(7046):1214-7. (PMID: 15988521) Nanoscale. 2017 Dec 7;9(47):18723-18730. (PMID: 29165498) IEEE Trans Med Imaging. 2010 Nov;29(11):1851-9. (PMID: 20529726) Nat Commun. 2019 Apr 26;10(1):1936. (PMID: 31028253) IEEE Trans Med Imaging. 2011 Jun;30(6):1284-92. (PMID: 21317081) IEEE Trans Med Imaging. 2013 Sep;32(9):1565-75. (PMID: 23568496) IEEE Trans Med Imaging. 2011 Sep;30(9):1581-90. (PMID: 21402508) Phys Med Biol. 2024 May 21;69(11):. (PMID: 38593815) ILIVER. 2022 Nov 08;1(4):237-244. (PMID: 40636343) Phys Med Biol. 2007 Nov 7;52(21):6363-74. (PMID: 17951848) IEEE Trans Med Imaging. 2022 Dec;41(12):3562-3574. (PMID: 35816533) Nat Biomed Eng. 2020 Mar;4(3):325-334. (PMID: 32015409) Med Phys. 2011 Mar;38(3):1619-26. (PMID: 21520874) Phys Med Biol. 2022 Jun 10;67(12):. (PMID: 35533677) Vis Comput Ind Biomed Art. 2022 Oct 1;5(1):24. (PMID: 36180612) Theranostics. 2022 Nov 2;12(17):7641. (PMID: 36438493)
Grant Information:	2025YFC2426200 National Key R&D Program of China; 82227802 National Natural Science Foundation of China; Taishan Industrial Experts Program
Contributed Indexing:	Keywords: Transformer; image reconstruction; magnetic particle imaging; resolution enhancement
Entry Date(s):	Date Created: 20260722 Date Completed: 20260722 Latest Revision: 20260813
Update Code:	20260814
PubMed Central ID:	PMC13389105
DOI:	10.1002/mp.70565
PMID:	42482365
Βάση Δεδομένων:	MEDLINE

Περιγραφή
ISSN:	2473-4209
DOI:	10.1002/mp.70565