CRUNet-MR-Univ: A Foundation Model for Diverse Cardiac MRI Reconstruction

2026-01-07

arXiv.org

AI Summary

This work addresses the limited generalization of existing deep learning methods in cardiac MRI reconstruction, which struggle to handle multidimensional variations such as image contrast, sampling patterns, scanner vendors, anatomical structures, and disease types. To overcome this challenge, the authors propose CRUNet-MR-Univ, a foundational model that uniquely integrates prompt-based conditioning with spatiotemporal correlation modeling into a unified reconstruction framework. By leveraging prompts to inject prior knowledge and fusing spatiotemporal features, the model effectively adapts to diverse clinical scenarios. Extensive experiments demonstrate that CRUNet-MR-Univ consistently outperforms current state-of-the-art methods across various acceleration factors and distribution shifts, exhibiting superior generalization capability and strong potential for real-world clinical deployment.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

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

Abstract

In recent years, deep learning has attracted increasing attention in the field of Cardiac MRI (CMR) reconstruction due to its superior performance over traditional methods, particularly in handling higher acceleration factors, highlighting its potential for real-world clinical applications. However, current deep learning methods remain limited in generalizability. CMR scans exhibit wide variability in image contrast, sampling patterns, scanner vendors, anatomical structures, and disease types. Most existing models are designed to handle only a single or narrow subset of these variations, leading to performance degradation when faced with distribution shifts. Therefore, it is beneficial to develop a unified model capable of generalizing across diverse CMR scenarios. To this end, we propose CRUNet-MR-Univ, a foundation model that leverages spatio-temporal correlations and prompt-based priors to effectively handle the full diversity of CMR scans. Our approach consistently outperforms baseline methods across a wide range of settings, highlighting its effectiveness and promise.

Problem

• Cardiac MRI reconstruction
• generalizability
• distribution shift
• image diversity
• deep learning

Innovation

• foundation model
• cardiac MRI reconstruction
• spatio-temporal correlation
• prompt-based prior
• generalization