Guiding Quantitative MRI Reconstruction with Phase-wise Uncertainty

2025-02-28

AI Summary

Quantitative MRI (qMRI) multi-phase undersampled reconstruction suffers from ill-posed inverse problems, and existing methods rarely leverage reconstruction uncertainty to improve parametric map accuracy. This paper proposes an uncertainty-driven two-stage paradigm: first, a deep learning network reconstructs multi-phase images while explicitly estimating phase-wise uncertainty—via Bayesian approximation or entropy-based estimation; second, the estimated uncertainty is incorporated as confidence-weighted priors into the physics-based model fitting step to yield information-weighted T1/T2 mapping. Crucially, this is the first approach to directly propagate phase-level reconstruction uncertainty into subsequent parameter estimation, thereby enhancing quantitative fidelity. Evaluated on healthy subject data, the method achieves state-of-the-art performance, with显著 reductions in T1 and T2 map errors. The implementation is publicly available.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Multiagent Systems: Multiagent Systems under Uncertainty

Application Category

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

Abstract

Quantitative magnetic resonance imaging (qMRI) requires multi-phase acqui-sition, often relying on reduced data sampling and reconstruction algorithms to accelerate scans, which inherently poses an ill-posed inverse problem. While many studies focus on measuring uncertainty during this process, few explore how to leverage it to enhance reconstruction performance. In this paper, we in-troduce PUQ, a novel approach that pioneers the use of uncertainty infor-mation for qMRI reconstruction. PUQ employs a two-stage reconstruction and parameter fitting framework, where phase-wise uncertainty is estimated during reconstruction and utilized in the fitting stage. This design allows uncertainty to reflect the reliability of different phases and guide information integration during parameter fitting. We evaluated PUQ on in vivo T1 and T2 mapping datasets from healthy subjects. Compared to existing qMRI reconstruction methods, PUQ achieved the state-of-the-art performance in parameter map-pings, demonstrating the effectiveness of uncertainty guidance. Our code is available at https://anonymous.4open.science/r/PUQ-75B2/.

Problem

• Enhances qMRI reconstruction using phase-wise uncertainty
• Solves ill-posed inverse problem in multi-phase qMRI
• Improves parameter mapping accuracy with uncertainty guidance

Innovation

• PUQ uses phase-wise uncertainty for qMRI reconstruction.
• Two-stage framework: reconstruction and parameter fitting.
• Uncertainty guides information integration in parameter fitting.