A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Lingjing, Zhang, Chengxiu, Yi, Yinqiao, Wang, Yida, Song, Yang, Yan, Xu, Xu, Shengfang, Zhu, Dalin, Cao, Mengqiu, Zhou, Yan, Wang, Chenglong, Yang, Guang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913984790134784
author Chen, Lingjing
Zhang, Chengxiu
Yi, Yinqiao
Wang, Yida
Song, Yang
Yan, Xu
Xu, Shengfang
Zhu, Dalin
Cao, Mengqiu
Zhou, Yan
Wang, Chenglong
Yang, Guang
author_facet Chen, Lingjing
Zhang, Chengxiu
Yi, Yinqiao
Wang, Yida
Song, Yang
Yan, Xu
Xu, Shengfang
Zhu, Dalin
Cao, Mengqiu
Zhou, Yan
Wang, Chenglong
Yang, Guang
contents We propose a deep learning-based approach that integrates MRI sequence parameters to improve the accuracy and generalizability of quantitative image synthesis from clinical weighted MRI. Our physics-driven neural network embeds MRI sequence parameters -- repetition time (TR), echo time (TE), and inversion time (TI) -- directly into the model via parameter embedding, enabling the network to learn the underlying physical principles of MRI signal formation. The model takes conventional T1-weighted, T2-weighted, and T2-FLAIR images as input and synthesizes T1, T2, and proton density (PD) quantitative maps. Trained on healthy brain MR images, it was evaluated on both internal and external test datasets. The proposed method achieved high performance with PSNR values exceeding 34 dB and SSIM values above 0.92 for all synthesized parameter maps. It outperformed conventional deep learning models in accuracy and robustness, including data with previously unseen brain structures and lesions. Notably, our model accurately synthesized quantitative maps for these unseen pathological regions, highlighting its superior generalization capability. Incorporating MRI sequence parameters via parameter embedding allows the neural network to better learn the physical characteristics of MR signals, significantly enhancing the performance and reliability of quantitative MRI synthesis. This method shows great potential for accelerating qMRI and improving its clinical utility.
format Preprint
id arxiv_https___arxiv_org_abs_2508_08123
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images
Chen, Lingjing
Zhang, Chengxiu
Yi, Yinqiao
Wang, Yida
Song, Yang
Yan, Xu
Xu, Shengfang
Zhu, Dalin
Cao, Mengqiu
Zhou, Yan
Wang, Chenglong
Yang, Guang
Image and Video Processing
Computer Vision and Pattern Recognition
We propose a deep learning-based approach that integrates MRI sequence parameters to improve the accuracy and generalizability of quantitative image synthesis from clinical weighted MRI. Our physics-driven neural network embeds MRI sequence parameters -- repetition time (TR), echo time (TE), and inversion time (TI) -- directly into the model via parameter embedding, enabling the network to learn the underlying physical principles of MRI signal formation. The model takes conventional T1-weighted, T2-weighted, and T2-FLAIR images as input and synthesizes T1, T2, and proton density (PD) quantitative maps. Trained on healthy brain MR images, it was evaluated on both internal and external test datasets. The proposed method achieved high performance with PSNR values exceeding 34 dB and SSIM values above 0.92 for all synthesized parameter maps. It outperformed conventional deep learning models in accuracy and robustness, including data with previously unseen brain structures and lesions. Notably, our model accurately synthesized quantitative maps for these unseen pathological regions, highlighting its superior generalization capability. Incorporating MRI sequence parameters via parameter embedding allows the neural network to better learn the physical characteristics of MR signals, significantly enhancing the performance and reliability of quantitative MRI synthesis. This method shows great potential for accelerating qMRI and improving its clinical utility.
title A Physics-Driven Neural Network with Parameter Embedding for Generating Quantitative MR Maps from Weighted Images
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.08123