Information-Theoretic Optimization for Task-Adapted Compressed Sensing Magnetic Resonance Imaging

2026-04-14

AI Summary

This work addresses the limitations of existing task-adaptive compressed sensing MRI methods, which struggle to flexibly support arbitrary sampling rates and diverse clinical tasks within an end-to-end framework and lack effective uncertainty modeling. The authors propose a unified, end-to-end trainable framework grounded in information theory, which jointly optimizes sampling, reconstruction, and task inference by maximizing the mutual information between undersampled k-space data and downstream tasks. Leveraging variational mutual information bounds, amortized optimization, and probabilistic inference models, the approach accommodates both joint reconstruction-and-task and task-only clinical scenarios while enabling privacy preservation. Experiments demonstrate that the method achieves Dice scores comparable to deterministic baselines yet significantly outperforms them in posterior distribution alignment, as evidenced by superior Generalized Energy Distance (GED) metrics.

Technology Category

• Machine Learning: Information Theory
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

Task-adapted compressed sensing magnetic resonance imaging (CS-MRI) is emerging to address the specific demands of downstream clinical tasks with significantly fewer k-space measurements than required by Nyquist sampling. However, existing task-adapted CS-MRI methods suffer from the uncertainty problem for medical diagnosis and cannot achieve adaptive sampling in end-to-end optimization with reconstruction or clinical tasks. To address these limitations, we propose the first task-adapted CS-MRI from the information-theoretic perspective to simultaneously achieve probabilistic inference for uncertainty prediction and adapt to arbitrary sampling ratios and versatile clinical applications. Specifically, we formalize the task-adapted CS-MRI optimization problem by maximizing the mutual information between undersampled k-space measurements and clinical tasks to enable probabilistic inference for addressing the uncertainty problem. We leverage amortized optimization and construct tractable variational bounds for mutual information to jointly optimize sampling, reconstruction, and task-inference models, which enables flexible sampling ratio control using a single end-to-end trained model. Furthermore, the proposed framework addresses two kinds of distinct clinical scenarios within a unified approach, i.e., i) joint task and reconstruction, where reconstruction serves as an auxiliary process to enhance task performance; and ii) task implementation with suppressed reconstruction, applicable for privacy protection. Extensive experiments on large-scale MRI datasets demonstrate that the proposed framework achieves highly competitive performance on standard metrics like Dice compared to deterministic counterpart but provides better distribution matching to the ground-truth posterior distribution as measured by the generalized energy distance (GED).

Problem

• task-adapted CS-MRI
• uncertainty problem
• adaptive sampling
• clinical tasks
• k-space measurements

Innovation

• information-theoretic optimization
• task-adapted CS-MRI
• mutual information maximization
• probabilistic inference
• amortized optimization