MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction

MRF-Mixer is a novel deep learning method for magnetic resonance fingerprinting (MRF) reconstruction, offering 200× faster processing (0.35 s on CPU and 0.3 ms on GPU) and 40% higher accuracy (lower MAE) than dictionary matching. It develops a simulation-driven approach using complex-valued multi-la...

Full description

Saved in:
Bibliographic Details
Main Authors: Tianyi Ding, Yang Gao, Zhuang Xiong, Feng Liu, Martijn A. Cloos, Hongfu Sun
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Information
Subjects:
Online Access:https://www.mdpi.com/2078-2489/16/3/218
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850280194592473088
author Tianyi Ding
Yang Gao
Zhuang Xiong
Feng Liu
Martijn A. Cloos
Hongfu Sun
author_facet Tianyi Ding
Yang Gao
Zhuang Xiong
Feng Liu
Martijn A. Cloos
Hongfu Sun
author_sort Tianyi Ding
collection DOAJ
description MRF-Mixer is a novel deep learning method for magnetic resonance fingerprinting (MRF) reconstruction, offering 200× faster processing (0.35 s on CPU and 0.3 ms on GPU) and 40% higher accuracy (lower MAE) than dictionary matching. It develops a simulation-driven approach using complex-valued multi-layer perceptrons and convolutional neural networks to efficiently process MRF data, enabling generalization across sequence and acquisition parameters and eliminating the need for extensive in vivo training data. Evaluation on simulated and in vivo data showed that MRF-Mixer outperforms dictionary matching and existing deep learning methods for T1 and T2 mapping. In six-shot simulations, it achieved the highest PSNR (T1: 33.48, T2: 35.9) and SSIM (T1: 0.98, T2: 0.98) and the lowest MAE (T1: 28.8, T2: 4.97) and RMSE (T1: 72.9, T2: 13.67). In vivo results further demonstrate that single-shot reconstructions using MRF-Mixer matched the quality of multi-shot acquisitions, highlighting its potential to reduce scan times. These findings suggest that MRF-Mixer enables faster, more accurate multiparametric tissue mapping, substantially improving quantitative MRI for clinical applications by reducing acquisition time while maintaining imaging quality.
format Article
id doaj-art-49b6aaa66b9c46f6b8c5254457fa187b
institution OA Journals
issn 2078-2489
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Information
spelling doaj-art-49b6aaa66b9c46f6b8c5254457fa187b2025-08-20T01:48:50ZengMDPI AGInformation2078-24892025-03-0116321810.3390/info16030218MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting ReconstructionTianyi Ding0Yang Gao1Zhuang Xiong2Feng Liu3Martijn A. Cloos4Hongfu Sun5School of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, AustraliaSchool of Computer Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, AustraliaSchool of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, AustraliaDonders Centre for Cognitive Neuroimaging, Radboud University, 6525 Nijmegen, The NetherlandsSchool of Electrical Engineering and Computer Science, The University of Queensland, Brisbane, QLD 4072, AustraliaMRF-Mixer is a novel deep learning method for magnetic resonance fingerprinting (MRF) reconstruction, offering 200× faster processing (0.35 s on CPU and 0.3 ms on GPU) and 40% higher accuracy (lower MAE) than dictionary matching. It develops a simulation-driven approach using complex-valued multi-layer perceptrons and convolutional neural networks to efficiently process MRF data, enabling generalization across sequence and acquisition parameters and eliminating the need for extensive in vivo training data. Evaluation on simulated and in vivo data showed that MRF-Mixer outperforms dictionary matching and existing deep learning methods for T1 and T2 mapping. In six-shot simulations, it achieved the highest PSNR (T1: 33.48, T2: 35.9) and SSIM (T1: 0.98, T2: 0.98) and the lowest MAE (T1: 28.8, T2: 4.97) and RMSE (T1: 72.9, T2: 13.67). In vivo results further demonstrate that single-shot reconstructions using MRF-Mixer matched the quality of multi-shot acquisitions, highlighting its potential to reduce scan times. These findings suggest that MRF-Mixer enables faster, more accurate multiparametric tissue mapping, substantially improving quantitative MRI for clinical applications by reducing acquisition time while maintaining imaging quality.https://www.mdpi.com/2078-2489/16/3/218deep learningMR fingerprintingself-supervisedMRF-Mixer
spellingShingle Tianyi Ding
Yang Gao
Zhuang Xiong
Feng Liu
Martijn A. Cloos
Hongfu Sun
MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
Information
deep learning
MR fingerprinting
self-supervised
MRF-Mixer
title MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
title_full MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
title_fullStr MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
title_full_unstemmed MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
title_short MRF-Mixer: A Simulation-Based Deep Learning Framework for Accelerated and Accurate Magnetic Resonance Fingerprinting Reconstruction
title_sort mrf mixer a simulation based deep learning framework for accelerated and accurate magnetic resonance fingerprinting reconstruction
topic deep learning
MR fingerprinting
self-supervised
MRF-Mixer
url https://www.mdpi.com/2078-2489/16/3/218
work_keys_str_mv AT tianyiding mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction
AT yanggao mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction
AT zhuangxiong mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction
AT fengliu mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction
AT martijnacloos mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction
AT hongfusun mrfmixerasimulationbaseddeeplearningframeworkforacceleratedandaccuratemagneticresonancefingerprintingreconstruction