TuLaBM: Tumor-Biased Latent Bridge Matching for Contrast-Enhanced MRI Synthesis

TL;DR

This paper introduces TuLaBM, a novel efficient method for synthesizing contrast-enhanced MRI from non-contrast MRI by modeling the translation as latent space Brownian bridge transport, improving tumor detail fidelity and computational efficiency.

Contribution

The paper proposes a new latent space bridge matching framework with tumor-biased attention and boundary loss, advancing MRI synthesis accuracy and speed over existing GAN and diffusion models.

Findings

Outperforms state-of-the-art methods on multiple datasets.

Achieves inference times under 0.097 seconds per image.

Generalizes well to unseen data in zero-shot and fine-tuning scenarios.

Abstract

Contrast-enhanced magnetic resonance imaging (CE-MRI) plays a crucial role in brain tumor assessment; however, its acquisition requires gadolinium-based contrast agents (GBCAs), which increase costs and raise safety concerns. Consequently, synthesizing CE-MRI from non-contrast MRI (NC-MRI) has emerged as a promising alternative. Early Generative Adversarial Network (GAN)-based approaches suffered from instability and mode collapse, while diffusion models, despite impressive synthesis quality, remain computationally expensive and often fail to faithfully reproduce critical tumor contrast patterns. To address these limitations, we propose Tumor-Biased Latent Bridge Matching (TuLaBM), which formulates NC-to-CE MRI translation as Brownian bridge transport between source and target distributions in a learned latent space, enabling efficient training and inference. To enhance tumor-region fidelity, we introduce a Tumor-Biased Attention Mechanism (TuBAM) that amplifies tumor-relevant latent features during bridge evolution, along with a boundary-aware loss that constrains tumor interfaces to improve margin sharpness. While bridge matching has been explored for medical image translation in pixel space, our latent formulation substantially reduces computational cost and inference time. Experiments on BraTS2023-GLI (BraSyn) and Cleveland Clinic (in-house) liver MRI dataset show that TuLaBM consistently outperforms state-of-the-art baselines on both whole-image and tumor-region metrics, generalizes effectively to unseen liver MRI data in zero-shot and fine-tuned settings, and achieves inference times under 0.097 seconds per image.