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

Source: [TuLaBM: Tumor-Biased Latent Bridge Matching for Contrast-Enhanced MRI Synthesis](<https://www.sota2.com/research/paper/69c0c9060636e5bf4afc3bb3>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-24T05:00:03.763Z

Authors: Atharva Rege, Adinath Madhavrao Dukre, Numan Balci, Dwarikanath Mahapatra, Imran Razzak

Published: 2026-03-19T18:22:09.000Z

## 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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [NC-MRI to CE-MRI Synthesis on Cleveland Clinic Liver MRI-in house](<https://www.sota2.com/research/sota/nc-mri-to-ce-mri-synthesis-on-cleveland-clinic-liver-mri-in-house.md>): NC-MRI to CE-MRI Synthesis; Cleveland Clinic Liver MRI-in house; PSNR: 19.34.

- [NC-MRI to CE-MRI Synthesis on BraTS GLI (BraSyn) 2023](<https://www.sota2.com/research/sota/nc-mri-to-ce-mri-synthesis-on-brats-gli-brasyn-2023.md>): NC-MRI to CE-MRI Synthesis; BraTS GLI (BraSyn) 2023; PSNR (Whole Image): 25.16.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.19386>)

[PDF](<https://arxiv.org/pdf/2603.19386.pdf>)
