CUTE-MRI: Conformalized Uncertainty-based framework for Time-adaptivE MRI

2025-08-20

AI Summary

MRI acceleration reconstruction is ill-posed, propagating uncertainty to downstream clinical tasks; conventional fixed-acceleration strategies fail to jointly optimize scanning efficiency and diagnostic reliability. This paper proposes an uncertainty-aware dynamic adaptive MRI acquisition framework: for the first time, conformal prediction is integrated into the MRI acquisition pipeline to enable end-to-end uncertainty quantification—from k-space sampling and probabilistic reconstruction to clinically relevant metrics (e.g., cartilage thickness, cardiac functional parameters). Based on real-time confidence interval assessment, scanning is terminated adaptively, providing statistically verifiable accuracy guarantees. Evaluated on knee and cardiac MRI datasets, the method significantly reduces scan time compared to fixed-acceleration baselines while strictly satisfying pre-specified diagnostic accuracy thresholds.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Reasoning under Uncertainty: Uncertainty Representations

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
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Magnetic Resonance Imaging (MRI) offers unparalleled soft-tissue contrast but is fundamentally limited by long acquisition times. While deep learning-based accelerated MRI can dramatically shorten scan times, the reconstruction from undersampled data introduces ambiguity resulting from an ill-posed problem with infinitely many possible solutions that propagates to downstream clinical tasks. This uncertainty is usually ignored during the acquisition process as acceleration factors are often fixed a priori, resulting in scans that are either unnecessarily long or of insufficient quality for a given clinical endpoint. This work introduces a dynamic, uncertainty-aware acquisition framework that adjusts scan time on a per-subject basis. Our method leverages a probabilistic reconstruction model to estimate image uncertainty, which is then propagated through a full analysis pipeline to a quantitative metric of interest (e.g., patellar cartilage volume or cardiac ejection fraction). We use conformal prediction to transform this uncertainty into a rigorous, calibrated confidence interval for the metric. During acquisition, the system iteratively samples k-space, updates the reconstruction, and evaluates the confidence interval. The scan terminates automatically once the uncertainty meets a user-predefined precision target. We validate our framework on both knee and cardiac MRI datasets. Our results demonstrate that this adaptive approach reduces scan times compared to fixed protocols while providing formal statistical guarantees on the precision of the final image. This framework moves beyond fixed acceleration factors, enabling patient-specific acquisitions that balance scan efficiency with diagnostic confidence, a critical step towards personalized and resource-efficient MRI.

Problem

• Reducing MRI scan times while maintaining diagnostic quality
• Addressing uncertainty in accelerated MRI reconstruction from undersampled data
• Providing patient-specific adaptive acquisitions with statistical precision guarantees

Innovation

• Dynamic uncertainty-aware acquisition framework
• Probabilistic reconstruction model estimates image uncertainty
• Conformal prediction provides calibrated confidence intervals