Pretext Task Adversarial Learning for Unpaired Low-field to Ultra High-field MRI Synthesis

2025-03-07

AI Summary

To address the scarcity of high-field MRI scanners—leading to insufficient training data—and the inherently low signal-to-noise ratio and spatial resolution of low-field MRI, this paper proposes an unpaired low-field-to-ultra-high-field MRI synthesis framework. The method introduces a novel three-stage architecture: (1) slice-level geometric-preserving contrastive alignment (SGP) to ensure anatomical consistency; (2) local structural reconstruction (LSC) to enhance fine-grained texture fidelity; and (3) pretext-task-guided adversarial training to improve overall generation quality. By integrating contrastive learning, local rotation/masked reconstruction, and task-driven supervision, the approach achieves both anatomical accuracy and structural detail under strict unpaired data constraints. Evaluated on public benchmarks, it achieves state-of-the-art performance (FID = 16.892, IS = 1.933, MS-SSIM = 0.324), significantly improving synthetic image quality and generalization capability in downstream segmentation tasks.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Given the scarcity and cost of high-field MRI, the synthesis of high-field MRI from low-field MRI holds significant potential when there is limited data for training downstream tasks (e.g. segmentation). Low-field MRI often suffers from a reduced signal-to-noise ratio (SNR) and spatial resolution compared to high-field MRI. However, synthesizing high-field MRI data presents challenges. These involve aligning image features across domains while preserving anatomical accuracy and enhancing fine details. To address these challenges, we propose a Pretext Task Adversarial (PTA) learning framework for high-field MRI synthesis from low-field MRI data. The framework comprises three processes: (1) The slice-wise gap perception (SGP) network aligns the slice inconsistencies of low-field and high-field datasets based on contrastive learning. (2) The local structure correction (LSC) network extracts local structures by restoring the locally rotated and masked images. (3) The pretext task-guided adversarial training process introduces additional supervision and incorporates a discriminator to improve image realism. Extensive experiments on low-field to ultra high-field task demonstrate the effectiveness of our method, achieving state-of-the-art performance (16.892 in FID, 1.933 in IS, and 0.324 in MS-SSIM). This enables the generation of high-quality high-field-like MRI data from low-field MRI data to augment training datasets for downstream tasks. The code is available at: https://github.com/Zhenxuan-Zhang/PTA4Unpaired_HF_MRI_SYN.

Problem

• Synthesize high-field MRI from low-field MRI data
• Address challenges in aligning image features across domains
• Enhance anatomical accuracy and fine details in MRI synthesis

Innovation

• Slice-wise gap perception network aligns inconsistencies
• Local structure correction network restores details
• Pretext task-guided adversarial training enhances realism