Bootstrapped Physically-Primed Neural Networks for Robust T2 Distribution Estimation in Low-SNR Pancreatic MRI

2026-03-14

AI Summary

This study addresses the severely ill-posed inverse problem and noise sensitivity inherent in multi-component T2 relaxation distribution estimation from low signal-to-noise ratio pancreatic MRI. To this end, the work introduces inference-time guided sampling into T2 quantitative imaging for the first time, constructing a physics-informed neural network based on the P2T2 architecture. By integrating non-negative least squares regularization with a bootstrap ensemble strategy, the method transforms deterministic predictions into probabilistic ensemble outputs, effectively modeling the distributional characteristics of input signals. The proposed approach significantly enhances estimation robustness and physiological plausibility, achieving the lowest Wasserstein distance across repeated scans and substantially outperforming conventional NNLS and deterministic deep learning methods in distinguishing type 1 diabetes patients from healthy subjects.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

Estimating multi-component T2 relaxation distributions from Multi-Echo Spin Echo (MESE) MRI is a severely ill-posed inverse problem, traditionally solved using regularized non-negative least squares (NNLS). In abdominal imaging, particularly the pancreas, low SNR and residual uncorrelated noise challenge classical solvers and deterministic deep learning models. We introduce a bootstrap-based inference framework for robust distributional T2 estimation that performs stochastic resampling of the echo train and aggregates predictions across multiple subsets. This treats the acquisition as a distribution rather than a fixed input, yielding variance-reduced, physically consistent estimates and converting deterministic relaxometry networks into probabilistic ensemble predictors. Applied to the P2T2 architecture, our method uses inference-time bootstrapping to smooth noise artifacts and enhance fidelity to the underlying relaxation distribution. Noninvasive pancreatic evaluation is limited by location and biopsy risks, highlighting the need for biomarkers capable of capturing early pathophysiological changes. In type 1 diabetes (T1DM), progressive beta-cell destruction begins years before overt hyperglycemia, yet current imaging cannot assess early islet decline. We evaluate clinical utility via a test-retest reproducibility study (N=7) and a T1DM versus healthy differentiation task (N=8). Our approach achieves the lowest Wasserstein distances across repeated scans and superior sensitivity to physiology-driven shifts in the relaxation-time distribution, outperforming NNLS and deterministic deep learning baselines. These results establish inference-time bootstrapping as an effective enhancement for quantitative T2 relaxometry in low-SNR abdominal imaging.

Problem

• T2 distribution estimation
• low-SNR MRI
• ill-posed inverse problem
• pancreatic imaging
• quantitative relaxometry

Innovation

• bootstrap inference
• T2 relaxometry
• physically-primed neural networks
• low-SNR MRI
• probabilistic ensemble prediction