Interactive Gadolinium-Free MRI Synthesis: A Transformer with Localization Prompt Learning

2025-03-03

AI Summary

Gadolinium-based contrast agents (GBCAs) pose safety risks in contrast-enhanced MRI (CE-MRI), necessitating gadolinium-free, high-fidelity synthetic enhancement. To address this, we propose the first clinically interactive, gadolinium-free CE-MRI synthesis framework. Our method introduces a localization prompt learning mechanism enabling radiologists to input diagnostic spatial prompts in real time; a hierarchical Transformer backbone integrating multi-stage local/global feature fusion and a blur-aware prompt generation module; and spatial–cross-attention collaborative modeling coupled with random feature perturbation for robust prompt optimization. Evaluated on multi-center clinical data, our approach significantly improves lesion contrast and diagnostic fidelity over state-of-the-art methods. Code is publicly available.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Prompt Engineering / Prompting
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Contrast-enhanced magnetic resonance imaging (CE-MRI) is crucial for tumor detection and diagnosis, but the use of gadolinium-based contrast agents (GBCAs) in clinical settings raises safety concerns due to potential health risks. To circumvent these issues while preserving diagnostic accuracy, we propose a novel Transformer with Localization Prompts (TLP) framework for synthesizing CE-MRI from non-contrast MR images. Our architecture introduces three key innovations: a hierarchical backbone that uses efficient Transformer to process multi-scale features; a multi-stage fusion system consisting of Local and Global Fusion modules that hierarchically integrate complementary information via spatial attention operations and cross-attention mechanisms, respectively; and a Fuzzy Prompt Generation (FPG) module that enhances the TLP model's generalization by emulating radiologists' manual annotation through stochastic feature perturbation. The framework uniquely enables interactive clinical integration by allowing radiologists to input diagnostic prompts during inference, synergizing artificial intelligence with medical expertise. This research establishes a new paradigm for contrast-free MRI synthesis while addressing critical clinical needs for safer diagnostic procedures. Codes are available at https://github.com/ChanghuiSu/TLP.

Problem

• Synthesizes CE-MRI without gadolinium-based contrast agents
• Enhances diagnostic accuracy with interactive clinical integration
• Introduces a Transformer framework with localization prompt learning

Innovation

• Transformer with Localization Prompts for MRI synthesis
• Hierarchical backbone with multi-scale feature processing
• Fuzzy Prompt Generation for enhanced model generalization