Generalizable 7T T1-map Synthesis from 1.5T and 3T T1 MRI with an Efficient Transformer Model

2025-07-11

AI Summary

The scarcity, high cost, and severe artifacts associated with 7T MRI severely hinder clinical adoption of T1 quantitative mapping. Method: We propose 7T-Restormer—a lightweight Transformer-based model that enables end-to-end synthesis of high-fidelity 7T-equivalent T1 maps directly from conventional 1.5T or 3T T1-weighted MRI. It employs a cross-field-strength unified modeling architecture and mixed-field-strength training to enhance generalizability, while maintaining only 10.5 million parameters for computational efficiency. Contribution/Results: Evaluated on multicenter multiple sclerosis (MS) data, 7T-Restormer reduces normalized mean square error (NMSE) by 64% (1.5T input) and 41% (3T input) versus ground-truth 7T T1 maps, significantly outperforming ResShift and ResViT in both PSNR and SSIM. This work establishes a scalable, cost-effective paradigm for acquiring high-field-strength T1 quantitative imaging without requiring 7T hardware.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Purpose: Ultra-high-field 7T MRI offers improved resolution and contrast over standard clinical field strengths (1.5T, 3T). However, 7T scanners are costly, scarce, and introduce additional challenges such as susceptibility artifacts. We propose an efficient transformer-based model (7T-Restormer) to synthesize 7T-quality T1-maps from routine 1.5T or 3T T1-weighted (T1W) images. Methods: Our model was validated on 35 1.5T and 108 3T T1w MRI paired with corresponding 7T T1 maps of patients with confirmed MS. A total of 141 patient cases (32,128 slices) were randomly divided into 105 (25; 80) training cases (19,204 slices), 19 (5; 14) validation cases (3,476 slices), and 17 (5; 14) test cases (3,145 slices) where (X; Y) denotes the patients with 1.5T and 3T T1W scans, respectively. The synthetic 7T T1 maps were compared against the ResViT and ResShift models. Results: The 7T-Restormer model achieved a PSNR of 26.0 +/- 4.6 dB, SSIM of 0.861 +/- 0.072, and NMSE of 0.019 +/- 0.011 for 1.5T inputs, and 25.9 +/- 4.9 dB, and 0.866 +/- 0.077 for 3T inputs, respectively. Using 10.5 M parameters, our model reduced NMSE by 64 % relative to 56.7M parameter ResShift (0.019 vs 0.052, p = <.001 and by 41 % relative to 70.4M parameter ResViT (0.019 vs 0.032, p = <.001) at 1.5T, with similar advantages at 3T (0.021 vs 0.060 and 0.033; p < .001). Training with a mixed 1.5 T + 3 T corpus was superior to single-field strategies. Restricting the model to 1.5T increased the 1.5T NMSE from 0.019 to 0.021 (p = 1.1E-3) while training solely on 3T resulted in lower performance on input 1.5T T1W MRI. Conclusion: We propose a novel method for predicting quantitative 7T MP2RAGE maps from 1.5T and 3T T1W scans with higher quality than existing state-of-the-art methods. Our approach makes the benefits of 7T MRI more accessible to standard clinical workflows.

Problem

• Synthesizing 7T-quality T1-maps from 1.5T/3T MRI scans
• Overcoming high cost and scarcity of 7T MRI scanners
• Improving image quality beyond current state-of-the-art methods

Innovation

• Transformer model synthesizes 7T T1-maps from lower-field MRI
• Efficient 7T-Restormer outperforms ResViT and ResShift models
• Mixed 1.5T and 3T training enhances synthesis performance