Bridging Synthetic-to-Real Gaps: Frequency-Aware Perturbation and Selection for Single-shot Multi-Parametric Mapping Reconstruction

2025-03-05

AI Summary

To address the domain gap between synthetic and real data in single-acquisition multiparametric quantitative MRI, this paper proposes the Frequency-aware Perturbation and Selection (FPS) framework—the first method enabling high-fidelity cross-domain reconstruction without ground-truth labels. Methodologically, FPS innovatively integrates Wasserstein-distance-modulated Frequency-aware Perturbation (WDFP) with a Hierarchical Frequency-aware Selection Network (HFSNet), establishing a perturbation–selection closed loop that jointly optimizes domain-invariant feature learning and dynamic selection of optimal reconstructions. Coupled with the MOLED ultra-fast acquisition paradigm, FPS enables unsupervised domain-adaptive reconstruction. Evaluated on real data from 5 healthy volunteers, 94 ischemic stroke patients, and 46 meningioma patients, FPS achieves average PSNR gains of 3.2 dB and SSIM improvements of 0.08 for T1, T2, and ADC parameter maps, and successfully generalizes to DTI. The code is publicly available.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Multi-modal Vision
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Data-centric artificial intelligence (AI) has remarkably advanced medical imaging, with emerging methods using synthetic data to address data scarcity while introducing synthetic-to-real gaps. Unsupervised domain adaptation (UDA) shows promise in ground truth-scarce tasks, but its application in reconstruction remains underexplored. Although multiple overlapping-echo detachment (MOLED) achieves ultra-fast multi-parametric reconstruction, extending its application to various clinical scenarios, the quality suffers from deficiency in mitigating the domain gap, difficulty in maintaining structural integrity, and inadequacy in ensuring mapping accuracy. To resolve these issues, we proposed frequency-aware perturbation and selection (FPS), comprising Wasserstein distance-modulated frequency-aware perturbation (WDFP) and hierarchical frequency-aware selection network (HFSNet), which integrates frequency-aware adaptive selection (FAS), compact FAS (cFAS) and feature-aware architecture integration (FAI). Specifically, perturbation activates domain-invariant feature learning within uncertainty, while selection refines optimal solutions within perturbation, establishing a robust and closed-loop learning pathway. Extensive experiments on synthetic data, along with diverse real clinical cases from 5 healthy volunteers, 94 ischemic stroke patients, and 46 meningioma patients, demonstrate the superiority and clinical applicability of FPS. Furthermore, FPS is applied to diffusion tensor imaging (DTI), underscoring its versatility and potential for broader medical applications. The code is available at https://github.com/flyannie/FPS.

Problem

• Addressing synthetic-to-real gaps in medical imaging reconstruction.
• Improving multi-parametric mapping accuracy and structural integrity.
• Enhancing domain adaptation for clinical applicability in imaging.

Innovation

• Frequency-aware perturbation and selection (FPS) technique
• Wasserstein distance-modulated frequency-aware perturbation (WDFP)
• Hierarchical frequency-aware selection network (HFSNet)